Data Augmentation with Background Correction for Robust Prediction Using Neural Networks

Abstract

To provide a data extension method and a system for robust prediction using a neural network.SOLUTION: A method used for generating a division mask and generating an object image containing image data representing an object comprises a step of integrating an object image into different backgrounds and using them for data extension in neural network training. The data expansion is performed by using hue adjustment of the object image and / or rendering 3D capture data corresponding to the object from a selection view. An inference score is analyzed to select the backgrounds of the images contained in a training dataset. The backgrounds are selected and the training image is repeatedly added to the training dataset during training. In addition, early or late fusion using object mask data is used for improving an inference performed by a neural network trained using the object mask data.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] When training a neural network to perform a predictive task such as object classification, the accuracy of the trained neural network is often limited by the quality of the training dataset. To train to yield a robust neural network, the network should be trained using challenging training images. For example, when training a neural network for hand pose recognition (e.g., thumbs up, peace sign, clenched fist, etc.), the network may have difficulty detecting the hand pose in front of certain environmental features. If the pose includes outstretched fingers, the network may perform well when the pose is in front of a nearly solid color environment, but may have difficulty when the pose is in front of an environment that includes certain color patterns. As another example, the network may have difficulty with certain poses from certain angles or when the environment and the hand are of similar hues. [Background technology]

[0002] However, whether a particular training image is difficult for a neural network may depend on many factors, such as the prediction task being performed, the architecture of the neural network, and other training images seen by the network. It is therefore difficult to construct a training data set that will result in a well-trained network by predicting which training images should be used to train the network. It may be possible to estimate what features of the training images may be difficult for a neural network. However, even when such an estimate is possible and accurate, it may be impossible or impractical to obtain enough images that exhibit those features to adequately train the network. Summary of the Invention [Means for solving the problem]

[0003] Embodiments of the present disclosure relate to data augmentation including background filtering for robust prediction using neural networks. Systems and methods are disclosed that provide data augmentation techniques, such as those based on background filtering, that can be used to increase the robustness of trained neural networks.

[0004] In contrast to conventional systems, the present disclosure performs correction of the background of the object to generate the training images. A segmentation mask may be generated and used to generate an object image including image data representing the object. The object image may be merged with a different background and used for data augmentation in training the neural network. Other aspects of the present disclosure provide data augmentation using hue adjustment (e.g., of the object image) and / or rendering three-dimensional capture data corresponding to the object from a selected view direction. The present disclosure also provides analysis of the inference scores to select backgrounds of images to be included in the training dataset. Backgrounds may be selected and training images may be added to the training dataset repeatedly during training (e.g., between epochs). Additionally, the present disclosure performs early or late fusion using object mask data to improve inferences performed by neural networks trained using the object mask data.

[0005] The present systems and methods for data augmentation including background filtering for robust prediction using neural networks are described in detail below with reference to the accompanying drawings. [Brief description of the drawings]

[0006] [Figure 1]1 is a data flow diagram illustrating an example process for training one or more machine learning models based at least on integrating an object image with a background, according to some embodiments of the present disclosure. [Diagram 2] 1 is a data flow diagram illustrating an example process for generating an object image and integrating the object image with one or more backgrounds, according to some embodiments of the present disclosure. [Diagram 3] FIG. 12 includes an example of pre-processing that may be used to generate an object mask that is used to generate an object image, according to some embodiments of the present disclosure. [Figure 4] FIG. 2 illustrates how a 3D capture of an object can be rasterized from multiple views, according to some embodiments of the present disclosure. [Figure 5A] 1 is a data flow diagram illustrating an example of inference using a machine learning model and early fusion of object mask data, according to some embodiments of the present disclosure. [Figure 5B] 1 is a data flow diagram illustrating an example of inference using a machine learning model and late-stage fusion of object mask data, according to some embodiments of the present disclosure. [Figure 6] 1 is a flow diagram illustrating a method for training one or more machine learning models based at least on integrating an object image with at least one background, according to some embodiments of the present disclosure. [Figure 7] 1 is a flow diagram illustrating a method for inference using a machine learning model, where an input corresponds to a mask of an image and at least a portion of an image, according to some embodiments of the present disclosure. [Figure 8] 1 is a flow diagram illustrating a method for selecting a background of an object for training one or more machine learning models, according to some embodiments of the present disclosure. [Figure 9] FIG. 1 is a block diagram of an example computing device suitable for use in implementing some embodiments of the present disclosure. [Figure 10] FIG. 1 is a block diagram of an exemplary data center suitable for use in implementing some embodiments of the present disclosure. [Figure 11A] 1 is an illustration of an example autonomous vehicle, according to some embodiments of the present disclosure. [Figure 11B] 11B is an illustration of camera positions and fields of view for the example autonomous vehicle of FIG. 11A, in accordance with some embodiments of the present disclosure. [Figure 11C] FIG. 11B is a block diagram of an example system architecture of the example autonomous vehicle of FIG. 11A in accordance with some embodiments of the present disclosure. [Figure 11D] FIG. 11B is a system diagram of communication between a cloud-based server and the example autonomous vehicle of FIG. 11A, according to some embodiments of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Systems and methods are disclosed for data augmentation including background filtering for robust prediction using neural networks.Embodiments of the present disclosure relate to data augmentation including background filtering for robust prediction using neural networks.Systems and methods are disclosed that provide data augmentation techniques, such as those based on background filtering, that can be used to increase the robustness of trained neural networks.

[0008] The disclosed embodiments may be implemented using a variety of different systems, such as automotive systems, robotics, aviation systems, vehicle systems, marine systems, smart area surveillance systems, simulation systems, and / or other technology fields. The disclosed techniques may be used for any perception-based or more generally image-based analysis using machine learning models, such as for monitoring and / or tracking objects and / or environments.

[0009] Applications of the disclosed techniques include multimodal sensor interfaces, which may be applied in the context of healthcare. For example, a patient who is intubated or otherwise unable to communicate through speech may use poses or gestures that are interpreted by a computing system. Applications of the disclosed techniques further include autonomous driving and / or vehicle control or interaction. For example, the disclosed techniques may be used to implement hand pose or gesture recognition for sensing in the cabin of the vehicle 1100 of FIGS. 11A-11D to control convenience features, e.g., control of multimedia options. Gesture pose recognition may also be applied to the environment external to the vehicle 1100 to control any of a variety of autonomous driving control operations, including advanced driver assistance system (ADAS) functions.

[0010] As various examples, the disclosed techniques may be implemented in systems that include or are included in one or more of a system for performing conversational AI or personal assistant operations, a system for performing simulation operations, a system for performing simulation operations for testing or validating autonomous machine applications, a system for performing deep learning operations, a system implemented using edge devices, a system incorporating one or more virtual machines (VMs), a system implemented at least partially in a data center, or a system that may be implemented at least partially using cloud computing resources.

[0011] In contrast to conventional systems, the present disclosure identifies regions in an image that correspond to an object and uses the regions to filter, remove, replace, or otherwise modify the object's background and / or the object itself to generate training images. According to the present disclosure, a segmentation mask may be generated that identifies one or more segments that correspond to the object and one or more segments that correspond to the object's background in the source image. The segmentation mask may be applied to the source image to identify the region that corresponds to the object, e.g., to generate an object image that includes image data representing the object. The object image may be integrated with a different background and used for data augmentation in training a neural network. Other aspects of the present disclosure perform data augmentation using hue adjustment (e.g., of the object image) and / or rendering three-dimensional captured data corresponding to the object from a selected view direction.

[0012] Further aspects of the present disclosure provide techniques for selecting object backgrounds for training a neural network. According to the present disclosure, a Machine Learning Model (MLM) may be at least partially trained, and inference data may be generated by the MLM using images including different backgrounds. The MLM may include the neural network being trained or a different MLM. Inference scores corresponding to the inference data may be analyzed to select one or more features of the training images, e.g., a particular background or background type of the images to be included in the training dataset. The images may be selected from existing images or may be generated using the object images and backgrounds using any suitable technique, e.g., those described herein. In at least one embodiment, one or more features may be selected and one or more corresponding training images may be added to the training dataset iteratively during training.

[0013] The present disclosure further provides techniques for using object mask data to improve inferences performed by a neural network trained using the object mask data. Late fusion may be performed, where one set of inference data is generated from a source image and another set of inference data is generated from an image capturing the object mask data, e.g., an object image (e.g., using two copies of the neural network). The sets of inference data may be fused and used to update the neural network. In a further example, early fusion may be performed, where the source image and an image capturing the object mask data are combined and inference data is generated from the combined image. The object mask data may be used to de-emphasize or otherwise modify the background of the source image.

[0014] Reference is now made to Figure 1, which is a data flow diagram illustrating an example process 100 for training one or more machine learning models based at least on integrating an object image with a background, according to some embodiments of the present disclosure. The process 100 is described, by way of example, with respect to a machine learning model (MLM) training system 140. Among other potential components, the MLM training system 140 may include a background integrator 102, an MLM trainer 104, an MLM post-processor 106, and a background selector 108.

[0015] At a high level, the process 100 may include a background integrator 102 that receives one or more of a background 110 (which may be referred to as a background image) and an object image 112 corresponding to one or more objects (e.g., to be classified, analyzed, and / or detected by the MLM 122). The background integrator 102 may integrate the object image 112 with the background 110 to generate image data that captures (e.g., represents) at least a portion of the background 110 and the object in the one or more images. The MLM trainer 104 may generate inputs 120 to one or more MLMs 122 from the image data. The MLM 122 may process the inputs 120 to generate one or more outputs 124. The MLM post-processor 106 may process the outputs 124 to generate predicted data 126 (e.g., inference scores, object class labels, object bounding boxes or shapes, etc.). The background selector 108 can analyze the prediction data 126 and select one or more of the backgrounds 110 and / or objects for training based on at least the prediction data 126. In some embodiments, the process 100 can be repeated for any number of iterations until one or more of the MLMs 122 are sufficiently trained, or the background selector 108 can be used once or intermittently to select the backgrounds 110 for the first training iteration and / or any other iterations.

[0016] For example, and without limitation, the MLM 122 described herein may include any type of machine learning model, such as machine learning models that use linear regression, logistic regression, decision trees, support vector machines (SVMs), naive Bayes, k-nearest neighbors (Knn), K-means clustering, random forests, dimensionality reduction algorithms, gradient boosting algorithms, neural networks (e.g., autoencoders, convolutional, recurrent, perceptrons, long / short term memory (LSTM), Hopfield, Boltzmann, deep belief, deconvolution, generative adversarial, liquid state machines, etc.), and / or other types of machine learning models.

[0017] The process 100 may be used, at least in part, to train one or more of the MLMs 122 to perform a predictive task. This disclosure focuses on pose recognition and / or gesture recognition, and more specifically hand pose recognition. However, the disclosed techniques are broadly applicable to training MLMs to perform a variety of possible predictive tasks, e.g., image and / or object classification tasks. Examples include object detection, bounding box or shape determination, object classification, pose classification, gesture classification, and many others. For example, FIG. 2 illustrates one example of an image 246 that may be captured by the input 120 to the MLM 122. The process 100 may be used to train the MLM 122 to predict a hand pose (e.g., thumbs up, thumbs down, fist, peace sign, palm, OK sign, etc.) depicted in the image 246.

[0018] In some examples, one or more of the iterations of process 100 may not include training one or more of the MLMs 122. For example, an iteration may be used by background selector 108 to select one or more of the backgrounds 110 and / or objects corresponding to object images 112 for inclusion in a training dataset used by MLM trainer 104 for training. Further, in some instances, process 100 may use one MLM 122 to select backgrounds 110 and / or objects for a training dataset in one iteration and train the same or a different MLM 122 using the training dataset (e.g., in a subsequent iteration of process 100 or otherwise). For example, an MLM 122 used to select from backgrounds 110 for training may be partially or fully trained to perform a predictive task. An iteration of process 100 uses the trained or partially trained MLM 122 to select another MLM 122 from the background 110 for training, and the iteration may be used to bootstrap the training of the other MLM 122 by selecting difficult background images and / or combinations of background and object images for training.

[0019] In various instances, one or more iterations of the process 100 may form a feedback loop, in which the background selector 108 uses the prediction data of the iteration (e.g., a training epoch) to select one or more of the backgrounds 110 and / or objects corresponding to the object images 112 for inclusion in a subsequent training dataset. In subsequent iterations (e.g., subsequent training epochs) of the process 100, the background integrator 102 may generate or otherwise prepare or select corresponding images that the MLM trainer 104 may incorporate into the training dataset. The training dataset may then be applied to the trained MLM 122 to generate prediction data 126 used by the background selector 108 to select one or more of the backgrounds 110 and / or objects corresponding to the object images 112 for inclusion in the subsequent training dataset. The feedback loop may be used to determine enhancements to ensure continuous improvement in the accuracy, generalization ability, and robustness of the trained MLM 122.

[0020] In various instances, based at least on the background selector 108 selecting one or more backgrounds and / or objects, the MLM trainer 104 incorporates one or more images from the background integrator 102 that include the selected background and / or a combination of the selected background and object. As one example, the MLM trainer 104 can add one or more images to a training dataset used for a previous training iteration and / or epoch. The training dataset can grow with each iteration. However, in some instances, the MLM trainer 104 can also remove one or more images from the training dataset used for a previous training iteration and / or epoch (e.g., based on a selection by the background selector 108 for removal and / or based on the training dataset exceeding a threshold number of images). In the illustrated example, the background integrator 102 can generate one or more images that will be included in the training dataset at the beginning of an iteration of the process 100 based on a selection made by the background selector 108. In other instances, one or more of the images may be pre-generated by background integrator 102, e.g., at least in part, prior to any training using process 100 and / or during one or more previous iterations. If the images are pre-generated, MLM trainer 104 may fetch the pre-generated images from storage based on a selection made by background selector 108.

[0021] Background selection as described herein may refer to selection of a background for inclusion in at least one image used for training. Background selection may also include selection of an object for inclusion in an image having a background. In at least one embodiment, background selector 108 may select a background and / or a combination of background and object based at least on the confidence of MLM 122 in one or more predictions made using MLM 122. In various instances, the confidence may be captured by a set of inference scores corresponding to predictions of a prediction task performed by one or more of MLMs 122 on one or more images. For example, the predictions may be made in a current iteration and / or one or more previous iterations of process 100. Inference scores may refer to scores that the MLM is trained to provide or is trained to provide with respect to a prediction task or portion thereof. In some instances, the inference scores may represent the confidence of the MLM with respect to one or more of the corresponding outputs 124 (e.g., tensor data) and / or may be used to determine or calculate confidence with respect to a prediction task. For example, the inference score may represent the confidence of the MLM 122 in an object detected in an image belonging to a target class (eg, the probability of the input 120 belonging to the target class).

[0022] The background selector 108 can select one or more of the backgrounds and / or background and object combinations based on the inference scores using a variety of possible approaches. In at least one embodiment, the background selector 108 can select one or more particular backgrounds and / or objects based at least on an analysis of inference scores corresponding to images including those elements. In at least one embodiment, the background selector 108 can select one or more backgrounds and / or objects having one or more other particular characteristics based at least on an analysis of inference scores corresponding to images including those elements having a particular class or type or one or more characteristics thereof (e.g., a particular background, a particular object, texture, color, lighting conditions, objects and / or overlays, hue, viewpoint, orientation, skin color, size, theme, included background elements, etc., each of which are described with respect to FIG. 4). For example, the background selector 108 can select at least one background including blinds and a gesture including an open palm based at least on an analysis of inference scores corresponding to images including those elements sharing those characteristics. As another example, the background selector 108 can select a particular background based at least on the inference scores of images that include that background. As an additional example, the background selector 108 can select a particular background and object of an object class (e.g., thumbs up, thumbs down, etc.) based at least on the inference scores of images that include that background and objects of the object class.

[0023] In some instances, the inference score may be evaluated by the background selector 108 based at least on the calculation of the confusion score. The confusion score may serve as a metric that quantifies the relative network confusion with respect to predictions made for one or more images having a particular set of features. If the confusion score exceeds a threshold (e.g., indicates sufficient confusion), the background selector 108 may select at least some elements having the set of features (e.g., background or a combination of background and objects) to modify the training dataset. In at least one embodiment, the background confusion may be based at least in part on some correct and inaccurate predictions made for images including elements having the set of features. For example, the confusion score may be calculated based at least on a ratio of correct predictions to inaccurate predictions. Additionally or alternatively, the confusion score may be calculated based at least on a difference in accuracy of predictions or inference scores for images including elements having the set of features (e.g., indicating that there is a high difference across target classes when a particular background or background type is used).

[0024] The background selector 108 can select one or more backgrounds and / or background and object combinations for inclusion in at least one image of the training dataset based at least on the selection of one or more distinct element features. For example, the background selector 108 can select one or more distinct element features based at least on the corresponding confusion scores. The background selector 108 can rank the distinct sets of features and select one or more sets for modification to the training dataset based on the ranking. As a non-limiting example, the background selector 108 can select the top N particular backgrounds or background and object combinations (or other sets of features), where N is an integer (e.g., for each confusion score above a threshold).

[0025] The background integrator 102 may select, obtain (e.g., from storage), and / or generate one or more images that satisfy the selection made by the background selector 108 to modify the training dataset. If the background integrator 102 generates an image from a selected background, the background integrator 102 may include the entire background image or one or more portions of the background in the image. For example, the background integrator 102 may sample an area (e.g., a rectangle sized based on the input 120 to the MLM 122) from the background using a random or non-random sampling technique. Thus, the image processed by the MLM 122 may include the entire background image or a region of the background image (e.g., the sampled area). Similarly, if the background integrator 102 generates an image from a selected object, the background integrator 102 may include the entire object image or a portion of the object image in the image.

[0026] In at least one embodiment, the background integrator 102 can generate one or more synthetic backgrounds. For example, synthetic backgrounds can be generated for cases where network biases and sensitivities are understood (e.g., empirically or intuitively from measurements of network performance). For example, a particular type of synthetic background (e.g., dots and stripes) can be generated for networks that are sensitive to those particular textures, and then the background selector selects that background type. One or more synthetic backgrounds can be generated prior to and / or during any training of the MLM 122 (e.g., between iterations and epochs). A variety of possible approaches can be used to generate synthetic backgrounds, based at least in part on, for example, rendering a three-dimensional virtual environment associated with the background type, algorithmically generating textures that include the selected patterns, modifying existing backgrounds or images, etc.

[0027] According to aspects of the present disclosure, an object image 112 may be extracted from one or more source images, and a background integrator 102 may use one or more of the backgrounds 110 to replace or modify the original background of the source image. Referring now to Figure 2, Figure 2 is a data flow diagram illustrating an example process 200 for generating an object image 212 and integrating the object image 212 with one or more of the backgrounds 110, according to some embodiments of the present disclosure.

[0028] The process 200 is described with respect to an object image extraction system 202, as an example, which may include a region identifier 204, a pre-processor 206, and an image data determiner 208, among other potential components.

[0029] In overview, in the process 200, the region identifier 204 may be configured to identify a region in the source image. For example, the region identifier 204 may identify a region 212A in the source image 220 corresponding to an object (e.g., a hand) with a background in the source image 220. The region identifier 204 may further generate a segment mask 222 including a segment 212B corresponding to the object based on the identification of the region 212A. The region identifier 204 may detect the location of the object to define an area 230 of the source image 220 and / or the segment mask 222. The pre-processor 206 may process at least a portion of the segment 212B in the area 230 of the segment mask 222 to produce an object mask 232. The image data determiner 208 may generate an object image 212 from the source image 220 using the object mask 232. The object image may then be provided to the background integrator 102 for integration with one or more backgrounds 110 (e.g., to superimpose or overlay the object onto the background image).

[0030] According to various embodiments, one or more of object images 112, e.g., object image 212, may be generated prior to and / or during training of MLM 122. For example, one or more of object images 112 may be generated (e.g., as shown in FIG. 2 ) and stored, and then retrieved as needed to generate input 120 in process 100. As another example, one or more of object images 112 may be generated during process 100, e.g., on the fly or as needed by background integrator 102. In some embodiments, object images 112 may be generated during process 100 on the fly, and then stored and / or reused in subsequent iterations of process 100 and / or later to train MLMs other than MLM 122.

[0031] As described herein, the region identifier 204 can identify a region 212A in the source image 220 that corresponds to an object (e.g., a hand) with a background in the source image 220. In the illustrated example, the region identifier 204 can also identify a region 210A in the source image 220 that corresponds to a background of the object. In other examples, the region identifier 204 can simply identify the region 212A.

[0032] In at least one embodiment, the region identifier 204 can identify the region 212A and at least determine a segment 212B of the source image 220 that corresponds to the object based at least on performing image segmentation on the source image. The image segmentation can be further used to identify the region 210A and determine a segment 212B of the source image 220 that corresponds to the background based at least on performing image segmentation on the source image 220. In at least one embodiment, the region identifier 204 can generate data representing a segment mask 222 from the source image 220, where the segment mask 222 indicates the segment 212B that corresponds to the object (white pixels in FIG. 2) and / or the segment 210B that corresponds to the background (black pixels in FIG. 2).

[0033] The region classifier 204 may be implemented in a variety of possible ways, for example, using AI-powered background removal. In at least one embodiment, the region classifier 204 includes one or more MLMs trained to classify or label individual or groups of pixels of an image. For example, the MLMs may be trained to identify foreground (e.g., corresponding to objects) and / or background in an image, and the image segments may correspond to the foreground and / or background. As a non-limiting example, the region classifier 204 may be implemented using RTX Greenscreen background removal technology by NVIDIA Corporation. In some instances, the MLMs may be trained to identify object types and label pixels accordingly.

[0034] In at least one embodiment, the region identifier 204 may include one or more object detectors, such as an object detector trained to detect an object (e.g., a hand) to be classified by the MLM 122. The object detector may be implemented using one or more MLMs trained to detect the object. The object detector may output data indicative of the location of the object, which may be used to define an area 230 of the source image 220 and / or segmentation mask 222 that contains the object. For example, the object detector may be trained to provide a bounding box or shape of the object, which may be used to define the area 230.

[0035] In the illustrated example, area 230 may be defined by enlarging a bounding box, although in other examples, a bounding box may be used as area 230. In this example, region identifier 204 may identify section 212B that corresponds to the object by applying source image 220 to the MLM. In other examples, region identifier 204 may apply area 230 to the MLM rather than (or, in some embodiments, in addition to) the source image. By applying source image 220 to the MLM, the MLM may have additional context not available in area 230 that may increase the accuracy of the MLM.

[0036] In embodiments in which area 230 is determined, pre-processor 206 may perform pre-processing based at least on area 230. For example, area 230 of segmentation mask 222 may be pre-processed by pre-processor 206 before being used by image data determiner 208. In at least one embodiment, pre-processor 206 may crop image data corresponding to area 230 from segmentation mask 222 and process the cropped image data to produce object mask 232. Pre-processor 206 may perform various types of pre-processing on area 230 that may improve the capabilities of image data determiner 208.

[0037] Referring now to Figure 3, Figure 3 includes an example of pre-processing that may be used to generate an object mask used to generate an object image, according to some embodiments of the present disclosure. By way of example, the pre-processor 206 may crop the segmentation mask 222 resulting in the object mask 300A. The pre-processor 206 may perform a dilation on the object mask 300A resulting in the object mask 300B. The pre-processor 206 may then blur the object mask 300B resulting in the object mask 232. The object mask 232 may then be used by the image data determiner 208 to generate the object image 212.

[0038] The pre-processor 206 can use dilation to expand the section 212B of the object mask 300A that corresponds to the object. For example, the section 212B can be dilated to the section 210B that corresponds to the background. In an embodiment, the pre-processor 206 can perform binary dilation. Other types of dilation, such as grayscale dilation, can be performed. As an example, the pre-processor 206 can first blur the object mask 300A and then perform grayscale dilation. Dilation can be beneficial to increase the robustness of the object mask 232 against errors in the section mask 222. For example, if the object includes a hand, the palm can sometimes be classified as being in the background. Dilation is one approach to correct this potential error. Other mask pre-processing techniques are within the scope of this disclosure, such as binary or grayscale erosion. As an example, erosion can be performed on the section 210B that corresponds to the background.

[0039] The pre-processor 206 can use blurring (e.g., Gaussian blurring) to aid the background integrator 102 in smoothing the transition between image data corresponding to the object in the object image 212 and image data corresponding to the background 110. Without smoothing the transition between the object and the background, areas corresponding to the edges of the object mask 232 can be sharp and artificial. Using blending techniques such as blurring and then applying the object mask can result in a more natural or realistic transition between the object and the background 110 in the image 246. Although masking has been described as being performed on the object mask prior to application of the mask, in other instances, similar or different image processing operations can be performed by the image data determiner 208 in the application of the object mask (e.g., to the source image 220).

[0040] Returning to FIG. 2 , the image data determiner 208 may generally generate the object image 212 using an object mask, e.g., the object mask 232. For example, the image data determiner 208 may use the object mask 232 to identify and / or extract regions 242 from the source image 220 that correspond to the object. In other embodiments, an object mask may not be used, and another technique may be used to identify and / or extract the regions 242. When using the object mask 232, the image data determiner 208 may multiply the object mask 232 with the source image 220 to obtain the object image 212 including image data representing the regions 242 (e.g., the foreground of the source image 220) that correspond to the object.

[0041] When integrating the object image 212 with the background 110, the background integrator 102 can use the object image 212 as a mask, and the inverse of the mask can be applied to the background 110 with the resulting image blended with the object image. For example, the background integrator 102 can perform alpha compositing between the object image 212 and the background 110. The blending of the object image 212 with the background integrator 102 can use a variety of possible blending techniques. In some implementations, the background integrator 102 can integrate the object image 212 with the background 110 using alpha blending. The alpha blending can zero out the background from the object image 212 when combining the object image 212 with the background 110, and the foreground pixels can be superimposed on the background 110 to generate the image 246, or the pixels can be weighted (e.g., from 0 to 1) when combining the image data from the object image 212 and the background 110 with a blur applied by the pre-processor 206 or otherwise. In at least one embodiment, the background integrator 102 can use one or more seamless blending techniques that aim to create a seamless boundary between the object in the image 246 and the background 110. Examples of seamless blending techniques include gradient domain blending, Laplacian pyramid blending, or Poisson blending.

[0042] Further examples of data augmentation techniques As described herein, process 200 may be used to augment a training data set used to train an MLM, e.g., MLM 122 using process 100. This disclosure provides further techniques that may be used to enhance a training data set. According to at least some embodiments, the hue of an object identified in a source image, e.g., source image 220, may be modified for data augmentation. As an example, if the object represents at least a portion of a person, the skin color, hair color, and / or other hues may be modified to augment the training data set. For example, the hue of one or more portions of a region corresponding to the object may be altered (e.g., uniformly or otherwise). In at least one embodiment, the hue may be selected randomly or non-randomly. In some cases, the hue may be selected based on an analysis of prediction data 126. For example, the hue may be used as a feature to select or generate one or more training images, as described herein (e.g., by background integrator 102).

[0043] Certain areas, such as the background or non-primary or minor sections of regions that tend to have consistent hues for different real-world variations of an object, may retain their original hue. For example, if the object is a car, the panels may be changed in hue while preserving the hues of the lights, bumpers, and tires. In at least one embodiment, background integrator 102 may perform the hue correction. For example, hue correction may be performed on one or more portions of an object represented in object image 212. In other instances, object image 212 or object mask 232 may be used (e.g., by image data determiner 208) to identify image data representing the object and to correct the hue in one or more regions of source image 220. These instances may not include background integrator 102.

[0044] According to at least some embodiments, the source image 220 may be rendered from various different views of an object in an environment for data augmentation. Referring now to FIG. 4, FIG. 4 is a diagram of how a three-dimensional (3D) capture 402 of an object may be rasterized from multiple views according to some embodiments of the present disclosure. In at least one embodiment, the object image extraction system 202 may select a view of an object in the environment. For example, the object image extraction system 202 may select from views 406A, 406B, 406C, or any view of an object in the environment 400. The object image extraction system 202 may then generate the source image 220 based at least on the rasterization of the 3D capture of the object in the environment from the views. For example, the three-dimensional (3D) capture 402 may include depth information captured by a physical or virtual depth-sensing camera in a physical or virtual environment (which may be different from the environment 400). In one or more embodiments, the 3D capture 402 may include a point cloud capturing at least a portion of the object and potential additional elements of the environment 400. For example, if the view 406A is selected, the object image extraction system 202 may rasterize the source image 220 from the view 406A of the camera 404 using at least the 3D capture 402. In at least one embodiment, the view may be selected randomly or non-randomly. In some cases, the view may be selected based on an analysis of the prediction data 126. For example, the view may be used (e.g., by the background integrator 102) as a feature to select or generate one or more training images as described herein (e.g., in combination with the object class). In at least one embodiment, the object may be rasterized from the view to generate the object image 112, which may then be integrated with one or more backgrounds using techniques described herein. In other instances, the object may be rasterized with the background 110 (a two-dimensional image) or with other 3D content of the environment 400 to form a background.

[0045] Example of inference using object masks As described herein, the object mask may be used for data augmentation in training the MLM 122, for example using process 100. In at least one embodiment, the MLM 122 trained using mask data may perform inference on images without leveraging the object mask. For example, the input 120 to the MLM 122 during deployment may correspond to one or more images captured by a camera. In such an instance, the object mask may be used simply for data augmentation. In other embodiments, the object mask may also be leveraged for inference. An example of how the object mask may be leveraged for inference is described with respect to FIGS. 5A and 5B.

[0046] 5A, which is a data flow diagram 500 illustrating one example of inference and early fusion of object mask data using the MLM 122, according to some embodiments of the present disclosure. In the example of FIG. 5A, the MLM 122 may be trained to perform inference on the image 506 while utilizing an object image 508 that corresponds to an object mask in the image 506. For example, the input 120 may be generated from a combination of the image 506 and the object image 508 and then provided to the MLM 122 (e.g., a neural network), which may generate an output 124 including inference data 510. If the MLM 122 includes a neural network, the inference data 510 may include tensor data from the neural network. Post-processing may be performed on the inference data 510 to generate prediction data 126.

[0047] Object image 508 is one example of object mask data that may be combined with image 506 for inference. If early fusion of object mask data is used for inference, as in data flow diagram 500, input 120 to MLM 122 may be generated during training as well (e.g., in process 100). In general, object mask data may capture information about the manner in which region identifier 204 generates a mask from a source image. By leveraging object mask data during training and inference, MLM 122 may learn to account for any errors or unnatural artifacts that may be produced by object mask generation. Object mask data may also capture information about the manner in which preprocessor 206 preprocesses object masks to capture any errors or unnatural artifacts that may be produced by or remain after preprocessing.

[0048] Object image 508 may be generated (e.g., during inference) using object image extraction system 202 similar to object image 212. Although object image 508 is shown in Figures 5A and 5B, in other instances, segmentation mask 222 and / or object mask 232 may be used in addition to or instead of object image 508 (before or after preprocessing by preprocessor 206).

[0049] Various techniques may be used to generate the input 120 from the combination of the image 506 and the object image 508 (or, more generally, the object mask data). In at least one embodiment, the image 506 and the object image 508 are provided as separate inputs 120 to the MLM 122. As a further illustration, the image 506 and the object image 508 may be combined to form a combined image, and the input 120 may be generated from the combined image. In at least one embodiment, the object mask data may be used to fade, de-emphasize, mark, indicate, highlight, or otherwise modify one or more portions of the image 506 that represent the background (as captured by the object mask data) relative to the object or foreground of the image 506. For example, the image 506 may be blended with the object image 508, resulting in the background of the image 506 being faded, blurred, or defocused (e.g., using a depth of field effect). When combining image 506 and object image 508, the weights used to determine the resulting pixel color may decrease (e.g., exponentially) with distance from the object as dictated by the object mask data (e.g., using a dip-forming effect).

[0050] Reference is now made to FIG. 5B, which is a data flow diagram 502 illustrating an example of late-stage fusion of inference and object mask data using MLM 122, in accordance with some embodiments of the present disclosure.

[0051] In the example of FIG. 5B, the MLM 122 can provide separate outputs 124 for the image 506 and the object image 508. The outputs 124 can include inference data 510A corresponding to the image 506 and inference data 510B corresponding to the object image 508. Additionally, the MLM 122 can include separate inputs 120 for the image 506 and the object image 508. For example, the MLM 122 can include multiple copies of the MLM (122) trained to perform inference on images, where one copy performs inference on the image 506 and generates inference data 510A, and another copy performs inference on the object image 508 and generates inference data 510B (e.g., in parallel). Post-processing can be performed on the inference data 510A and the inference data 510B, and late fusion can be used to generate the prediction data 126. For example, corresponding tensor values ​​across inferred data 510A and inferred data 510B may be combined (e.g., averaged) to fuse the inferred data, and then further post-processing may be performed on the fused inferred data to generate predicted data 126. In at least one embodiment, the tensor values ​​of inferred data 510A and inferred data 510B may be combined using weights (e.g., using a weighted average). In at least one embodiment, the weights may be adjusted via a validation data set.

[0052] Inference using the MLM 122 may also include temporal filtering of inference scores to generate prediction data 126, which may improve the stability of predictions over time. In addition, the examples shown primarily relate to recognizing static poses. However, the disclosed techniques may also be applied to recognizing dynamic poses, which may be referred to as gestures. To train and use the MLM to predict gestures, in at least one embodiment, multiple images may be provided to the MLM 122 capturing an object over a period of time or over several or a series of frames. If object mask data is used, the object mask data may be provided for each input image.

[0053] Referring now to FIG. 6, each block of method 600, and other methods described herein, includes computational processes that may be performed using any combination of hardware, firmware, and / or software. For example, various functions may be performed by a processor executing instructions stored in a memory. The method may also be implemented as computer usable instructions stored on a computer storage medium. The method may be provided by a standalone application, a service, or a hosted service (standalone or in combination with another hosted service), or a plug-in to another product, to name a few. Additionally, method 600 is described with respect to system 140 of FIG. 1 and system 202 of FIG. 2, by way of example. However, the method may additionally or alternatively be performed by any one system or any combination of systems, including but not limited to those described herein.

[0054] 6 is a flow diagram illustrating a method 600 for training one or more machine learning models based at least on integrating an object image with at least one background, according to some embodiments of the present disclosure. The method 600 includes identifying a region in a first image corresponding to an object having a first background, at block B602. For example, the region identifier 204 can identify a region 212A in the source image 220 corresponding to an object having a background in the source image 220.

[0055] The method 600 includes determining image data representative of the object based at least on the region at block B604. For example, the image data determiner 208 can determine image data representative of the object based at least on the region 212A of the object. In at least one embodiment, the image data determiner 208 can determine the image data using the object mask 232 or a non-mask based approach.

[0056] The method 600 includes generating a second image including the object with a second background using the image data at block B606. The background integrator 102 can generate the image 246 including the object with the background 110 based at least on integrating the object with the background 110 using the image data. For example, the image data determiner 208 can incorporate the image data into the object image 212 and provide the object image 212 to the background integrator 102 for integration with the background 110.

[0057] The method 600 includes training at least one neural network to perform a predictive task using the second image at block B608. For example, the MLM trainer 104 can train the MLM to classify objects in the image using the image 246.

[0058] Referring now to Figure 7, Figure 7 is a flow diagram illustrating a method 700 for inference using a machine learning model, where input corresponds to a mask of an image and at least a portion of the image, according to some embodiments of the present disclosure. The method 700 includes, at block B702, obtaining (or accessing) at least one neural network trained to perform a predictive task on an image using input generated from a mask corresponding to an object. For example, the MLM 122 of Figure 5A or 5B may be obtained (accessed) and may have been trained according to process 100 of Figure 1.

[0059] The method 700 includes generating a mask corresponding to an object in the image, where the object has a background in the image, at block B704. For example, the region identifier 204 can generate a segmentation mask 222 corresponding to an object in the source image 220, where the object has a background in the source image 220.

[0060] The method 700 includes generating an input to at least one neural network using the mask at block B706. For example, the input 120 of FIG. 5A or 5B may be generated using the segmentation mask 222 (or without using object mask data). The input 120 may capture an object having at least a portion of a background.

[0061] The method 700 includes generating at least one prediction of the prediction task based at least on application of the inputs to the at least one neural network at block B708. For example, the MLM 122 may be used to generate at least one prediction of the prediction task based at least on application of the inputs 120 to the MLM 122, and prediction data 126 may be determined using the output 124 from the MLM 122.

[0062] 8, which is a flow diagram illustrating a method 800 for selecting a background of an object for training one or more machine learning models, according to some embodiments of the present disclosure. The method 800 includes, at block B802, receiving an image of one or more objects having multiple backgrounds. For example, the MLM training system 140 can receive an image of one or more objects having multiple backgrounds 110.

[0063] The method 800 includes generating a set of reasoning scores corresponding to the prediction task using the images at block B 804. For example, the MLM trainer 104 can provide inputs 120 to one or more of the MLMs 122 (or different MLMs) to generate outputs 124, and the MLM post-processor 106 can process the outputs 124 to produce prediction data 126.

[0064] The method 800 includes selecting a background based at least on one or more of the inference scores at block B806. For example, the background selector 108 may select one or more of the backgrounds 110 based at least on the prediction data 126.

[0065] The method 800 includes generating an image based at least on integrating the object with the background at block B808. For example, the background integrator 102 can generate an image based at least on integrating the object with the background (e.g., using the object image 112 and the background 110).

[0066] The method 800 includes training at least one neural network using the images to perform a predictive task at block B 810. For example, the MLM trainer 104 can train one or more of the MLMs 122 using the images.

[0067] Exemplary Computing Device 9 is a block diagram of an example computing device 900 suitable for use in implementing some embodiments of the present disclosure. The computing device 900 may include an interconnect system 902 that directly or indirectly couples the following devices: memory 904, one or more central processing units (CPUs) 906, one or more graphics processing units (GPUs) 908, a communication interface 910, input / output (I / O) ports 912, input / output components 914, a power supply 916, one or more presentation components 918 (e.g., displays), and one or more logic units 920. In at least one embodiment, the computing device 900 may include one or more virtual machines (VMs) and / or any of its components may include virtual components (e.g., virtual hardware components). As non-limiting examples, one or more of GPUs 908 may include one or more vGPUs, one or more of CPUs 906 may include one or more vCPUs, and / or one or more of logic units 920 may include one or more virtual logic units. As such, computing device 900 may include discrete components (e.g., an entire GPU dedicated to computing device 900), virtual components (e.g., a portion of a GPU dedicated to computing device 900), or a combination thereof.

[0068] While the various blocks in FIG. 9 are shown as connected by lines via the interconnect system 902, this is not intended to be limiting and is merely for clarity. For example, in some embodiments, the presentation component 918, e.g., a display device, may be considered an I / O component 914 (e.g., where the display is a touch screen). As another example, the CPU 906 and / or the GPU 908 may include memory (e.g., the memory 904 may represent a storage device in addition to the memory of the GPU 908, the CPU 906, and / or other components). In other words, the computing devices of FIG. 9 are merely exemplary. Categories such as "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "gaming consoles," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types are all intended to be within the scope of the computing devices of FIG. 9 and therefore are not to be differentiated.

[0069] The interconnect system 902 may represent one or more links or buses, such as an address bus, a data bus, a control bus, or a combination thereof. The interconnect system 902 may include one or more bus or link types, such as an industry standard architecture (ISA) bus, an extended industry standard architecture (EISA) bus, a video electronics standards association (VESA) bus, a peripheral component interconnect (PCI) bus, a peripheral component interconnect express (PCIe) bus, and / or another type of bus or link. In some embodiments, there are direct connections between components. As an example, the CPU 906 may be directly connected to the memory 904. Additionally, the CPU 906 may be directly connected to the GPU 908. When there are direct or point-to-point connections between components, the interconnect system 902 may include a PCIe link to implement the connections. In these examples, a PCI bus need not be included in the computing device 900.

[0070] Memory 904 may include any of a variety of computer readable media. Computer readable media may be any available media that can be accessed by computing device 900. Computer readable media may include both volatile and nonvolatile media, and removable and non-removable media. By way of example, and not limitation, computer readable media may comprise computer storage media and communication media.

[0071] Computer storage media may include both volatile and nonvolatile media and / or removable and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, and / or other data types. For example, memory 904 may store computer readable instructions (e.g., representing programs and / or program elements), such as an operating system. Computer storage media may include, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that may be used to store the desired information and that may be accessed by computing device 900. As used herein, computer storage media does not include the signals themselves.

[0072] Computer storage media may embody computer readable instructions, data structures, program modules, and / or other data types in a modulated data signal such as a carrier wave or other transport mechanism and include any information delivery media. The term "modulated data signal" may refer to a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, computer storage media may include wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared and other wireless media. Combinations of any of the above should also be included within the scope of computer readable media.

[0073] The CPU 906 may be configured to execute at least some of the computer readable instructions to control one or more components of the computing device 900 to perform one or more of the methods and / or processes described herein. The CPU 906 may include one or more (e.g., 1, 2, 4, 8, 28, 72, etc.) cores each capable of simultaneously processing multiple software threads. The CPU 906 may include any type of processor, and may include different types of processors depending on the type of computing device 900 implemented (e.g., a processor with fewer cores for a mobile device and a processor with more cores for a server). For example, depending on the type of computing device 900, the processor may be an Advanced RISC Machines (ARM) processor implemented using Reduced Instruction Set Computing (RISC) or an x86 processor implemented using Complex Instruction Set Computing (CISC). Computing device 900 may include one or more CPUs 906 in addition to one or more microprocessors or auxiliary coprocessors, such as computational coprocessors.

[0074] In addition to or instead of the CPU 906, the GPU 908 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 900 to perform one or more of the methods and / or processes described herein. One or more of the GPUs 908 may be integrated GPUs (e.g., with one or more of the CPUs 906) and / or one or more of the GPUs 908 may be discrete GPUs. In an embodiment, one or more of the GPUs 908 may be coprocessors of one or more of the CPUs 906. The GPU 908 may be used by the computing device 900 to render graphics (e.g., 3D graphics) or perform general-purpose computing. For example, the GPU 908 may be used for general-purpose computing on GPU (GPGPU). The GPU 908 may include hundreds or thousands of cores capable of processing hundreds or thousands of software threads simultaneously. GPU 908 may generate pixel data for an output image in response to a rendering command (e.g., a rendering command from CPU 906 received via a host interface). GPU 908 may include graphics memory, e.g., display memory, for storing pixel data or any other suitable data, e.g., GPGPU data. The display memory may be included as part of memory 904. The display memory may be included as part of memory 904. GPU 908 may include two or more GPUs operating in parallel (e.g., via links). The links may connect the GPUs directly (e.g., using NVLINK) or may connect the GPUs via a switch (e.g., using NVSwitch). When coupled together, each GPU 908 may generate pixel data or GPGPU data for a different portion of the output or for a different output (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU may include its own memory or may share memory with the other GPUs.

[0075] In addition to or instead of the CPU 906 and / or GPU 908, the logic unit 920 may be configured to execute at least some of the computer readable instructions to control one or more components of the computing device 900 to perform one or more of the methods and / or processes described herein. In an embodiment, the CPU 906, the GPU 908, and / or the logic unit 920 may execute any combination of the methods, processes and / or portions thereof, either discretely or jointly. One or more of the logic units 920 may be part of and / or integrated with one or more of the CPU 906 and / or GPU 908, and / or one or more of the logic units 920 may be a discrete component to or otherwise external to the CPU 906 and / or GPU 908. In an embodiment, one or more of the logic units 920 may be a co-processor of one or more of the CPU 906 and / or GPU 908.

[0076] Examples of logic unit 920 include one or more processing cores and / or components thereof, such as, for example, a Tensor Core (TC), a Tensor Processing Unit (TPU), a Pixel Visual Core (PVC), a Vision Processing Unit (VPU), a Graphics Processing Cluster (GPC), a Texture Processing Cluster (TPC), a Streaming Multiprocessor (SM), a Tree Traversal Unit (TTU), an Artificial Intelligence Accelerator (AIA), a Deep Learning Accelerator (DLA), an Arithmetic-Logic Unit (ALU), an Application Specific Integrated Circuit (ASIC), a Floating Point Unit (FPU), an Input / Output (I / O) element, a Peripheral Component Interconnect (PCI) or Peripheral Component Interconnect Express (PCIe) element, and / or the like.

[0077] The communications interface 910 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 900 to communicate with other computing devices over electronic communications networks, including wired and / or wireless communications. The communications interface 910 may include components and functionality to enable communication over any of a number of different networks, such as a wireless network (e.g., Wi-Fi, Z-Wave, Bluetooth, Bluetooth LE, ZigBee, etc.), a wired network (e.g., communicating over Ethernet or InfiniBand), a low power wide area network (e.g., LoRaWAN, SigFox, etc.), and / or the Internet.

[0078] The I / O ports 912 may enable the computing device 900 to be logically coupled to other devices, including I / O components 914, presentation components 918, and / or other components, some of which may be built-in (e.g., integrated) to the computing device 900. Exemplary I / O components 914 include a microphone, a mouse, a keyboard, a joystick, a game pad, a game controller, a satellite dish, a scanner, a printer, a wireless device, and the like. The I / O components 914 may provide a natural user interface (NUI) that processes air gestures, voice, or other physiological input generated by a user. In some cases, the input may be sent to an appropriate network element for further processing. The NUI may implement any combination of voice recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition on and next to the screen, air gestures, head and eye tracking, and touch recognition in conjunction with the display of the computing device 900 (as described in more detail below). The computing device 900 may include a depth camera for gesture detection and recognition, e.g., a stereoscopic camera system, an infrared camera system, an RGB camera system, touch screen technology, and combinations thereof. Additionally, the computing device 900 may include an accelerometer or gyroscope (e.g., as part of an inertia measurement unit (IMU)) to enable detection of motion. In some instances, the output of the accelerometer or gyroscope may be used by the computing device 900 to render immersive augmented or virtual reality.

[0079] The power supply 916 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 916 may provide power to the computing device 900 to enable the components of the computing device 900 to operate.

[0080] The presentation component 918 may include a display (e.g., a monitor, a touch screen, a television screen, a heads-up display (HUD), other display types, or a combination thereof), speakers, and / or other presentation components. The presentation component 918 can receive data from other components (e.g., GPU 908, CPU 906, etc.) and output data (e.g., as images, video, sound, etc.).

[0081] Exemplary Data Center 10 illustrates an example data center 1000 that may be used in at least one embodiment of the present disclosure. The data center 1000 may include a data center infrastructure layer 1010, a framework layer 1020, a software layer 1030, and / or an application layer 1040.

[0082] 10, the data center infrastructure layer 1010 may include a resource orchestrator 1012, grouped computational resources 1014, and node computational resources ("node CRs") 1016(1)-1016(N), where "N" represents any integer, natural number. In at least one embodiment, the node CRs 1016(1)-1016(N) may include, but are not limited to, any number of central processing units ("CPUs") or other processors (including accelerators, field programmable gate arrays (FPGAs), graphics processors or graphics processing units (GPUs), etc.), memory devices (e.g., dynamic read-only memory), storage devices (e.g., solid state or disk drives), network input / output ("NW I / O") devices, network switches, virtual machines ("VMs"), power modules, and / or cooling modules, etc. In some embodiments, one or more of the nodes CR 1016(1)-1016(N) may correspond to a server having one or more of the aforementioned computing resources. Additionally, in some embodiments, the nodes CR 1016(1)-1016(N) may include one or more virtual components, such as a vGPU, a vCPU, and / or the like, and / or one or more of the nodes CR 1016(1)-1016(N) may correspond to a virtual machine (VM).

[0083] In at least one embodiment, the grouped computing resources 1014 may include separate groups of nodes CR 1016 housed in one or more racks (not shown), or multiple racks housed in data centers in various geographic locations (also not shown). The separate groups of nodes CR 1016 in the grouped computing resources 1014 may include grouped computing, network, memory, or storage resources that may be configured or assigned to support one or more workloads. In at least one embodiment, several nodes CR 1016 including CPUs, GPUs, and / or other processors may be grouped in one or more racks to provide computing resources to support one or more workloads. The one or more racks may also include any number of power modules, cooling modules, and / or network switches in any combination.

[0084] The resource orchestrator 1022 may configure or otherwise control one or more nodes CR 1016(1)-1016(N) and / or grouped computational resources 1014. In at least one embodiment, the resource orchestrator 1022 may include a software design infrastructure ("SDI") management entity of the data center 1000. The resource orchestrator 1022 may include hardware, software, or some combination thereof.

[0085] In at least one embodiment, as shown in FIG. 10 , framework layer 1020 may include a job scheduler 1032, a configuration manager 1034, a resource manager 1036, and / or a distributed file system 1038. Framework layer 1020 may include frameworks to support software 1032 in software layer 1030 and / or one or more applications 1042 in application layer 1040. Software 1032 or applications 1042 may include web-based service software or applications, such as those offered by Amazon Web Services, Google Cloud, and Microsoft Azure, respectively. Framework layer 1020 may be, but is not limited to, a type of free and open source software web application framework, such as Apache Spark™ (hereinafter “Spark”), which may use a distributed file system 1038 for large-scale data processing (e.g., “big data”). In at least one embodiment, the job scheduler 1032 may include a Spark driver to facilitate scheduling of workloads supported by various tiers of the data center 1000. The configuration manager 1034 may be capable of configuring different tiers, for example, the software tier 1030 and the framework tier 1020 including Spark and a distributed file system 1038 to support large scale data processing. The resource manager 1036 may be capable of managing clustered or grouped computing resources that are mapped or assigned in support of the distributed file system 1038 and the job scheduler 1032. In at least one embodiment, the clustered or grouped computing resources may include the computing resources 1014 grouped in the data center infrastructure tier 1010. The resource manager 1036 may coordinate with the resource orchestrator 1012 to manage these mapped or assigned computing resources.

[0086] In at least one embodiment, the software 1032 included in the software layer 1030 may include software used by at least a portion of the nodes CRs 1016(1)-1016(N), the grouped computational resources 1014, and / or the distributed file system 1038 of the framework layer 1020. The one or more types of software may include, but are not limited to, Internet web page searching software, email virus scanning software, database software, and streaming video content software.

[0087] In at least one embodiment, the applications 1042 included in the application layer 1040 may include one or more types of applications used by at least a portion of the nodes CRs 1016(1)-1016(N), the grouped computational resources 1014, and / or the distributed file system 1038 of the framework layer 1020. The one or more types of applications may include, but are not limited to, any number of genomics applications, cognitive computing, and machine learning applications, including training or inference software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and / or other machine learning applications used in conjunction with one or more embodiments.

[0088] In at least one embodiment, any of the configuration manager 1034, resource manager 1036, and resource orchestrator 1012 can implement any number and type of self-rewriting actions based on any amount and type of data obtained in any technically possible manner. The self-rewriting actions can free the data center operator of the data center 1000 from making potentially poor configuration decisions and possibly avoiding underutilized and / or underperforming portions of the data center.

[0089] Data center 1000 may include tools, services, software, or other resources to train one or more machine learning models or to predict or infer information using one or more machine learning models according to one or more embodiments described herein. For example, a machine learning model may be trained by calculation of weight parameters via a neural network architecture using the software and / or computing resources described above with respect to data center 1000. In at least one embodiment, a trained or deployed machine learning model corresponding to one or more neural networks may be used to infer or predict information using the resources described above with respect to data center 1000, for example, by using weight parameters calculated via one or more training techniques, including but not limited to those described herein.

[0090] In at least one embodiment, data center 1000 may use CPUs, application specific integrated circuits (ASICs), GPUs, FPGAs, and / or other hardware (or corresponding virtual computing resources) for training and / or performing inference using such resources. Additionally, one or more of such software and / or hardware resources may be configured as services, such as image recognition, speech recognition, or other artificial intelligence services, to enable a user to train or perform inference on information.

[0091] Example Network Environment A network environment suitable for use in implementing embodiments of the present disclosure may include one or more client devices, servers, network attached storage (NAS), other back-end devices, and / or other device types. The client devices, servers, and / or other device types (e.g., each device) may be implemented with one or more instances of the computing device 900 of FIG. 9, e.g., each device may include similar components, features, and / or functionality of the computing device 900. In addition, if a back-end device (e.g., server, NAS, etc.) is implemented, the back-end device may be included as part of the data center 1000, examples of which are further detailed herein with respect to FIG. 10.

[0092] Components of a network environment may communicate with each other via a network, which may be wired, wireless, or both. A network may include multiple networks or a network of networks. Illustratively, a network may include one or more wide area networks (WANs), one or more local area networks (LANs), one or more public networks, such as the Internet and / or the Public Switched Telephone Network (PSTN), and / or one or more private networks. When a network includes a wireless telecommunications network, components such as base stations, communication towers, or access points (as well as other components) may provide wireless connectivity.

[0093] Compatible network environments may include one or more peer-to-peer network environments (wherein a server may not be included in the network environment) and one or more client-server network environments (wherein a server or servers may be included in the network environment). In a peer-to-peer network environment, the functionality described herein with respect to a server may be implemented in any number of client devices.

[0094] In at least one embodiment, the network environment may include one or more cloud-based network environments, distributed computing environments, combinations thereof, and the like. The cloud-based network environment may include a framework layer, a job scheduler, a resource manager, and a distributed file system implemented on one or more of the servers, which may include one or more core network servers and / or edge servers. The framework layer may include a framework to support software in the software layer and / or one or more applications in the application layer. The software or applications may include web-based service software or applications, respectively. In an embodiment, one or more of the client devices may use the web-based service software or applications (e.g., by accessing the service software and / or applications via one or more application programming interfaces (APIs)). The framework layer may be, but is not limited to, a type of free and open source software web application framework that may use a distributed file system for large-scale data processing (e.g., "big data"), for example.

[0095] A cloud-based network environment may provide cloud computing and / or cloud storage performing any combination of the computing and / or data storage functions (or one or more portions thereof) described herein. Any of these various functions may be distributed across multiple locations from a central or core server (e.g., one or more data centers that may be distributed across a state, territory, country, or world). When a connection to a user (e.g., a client device) is relatively close to an edge server, the core server may delegate at least a portion of the functionality to the edge server. A cloud-based network environment may be private (e.g., restricted to a single organization), public (e.g., available to multiple organizations), and / or a combination thereof (e.g., a hybrid cloud environment).

[0096] A client device may include at least some of the components, features, and functionality of the exemplary computing device 900 described herein with respect to Figure 9. By way of illustration, and without limitation, a client device may be embodied as a personal computer (PC), a laptop computer, a mobile device, a smartphone, a tablet computer, a smart watch, a wearable computer, a personal digital assistant (PDA), an MP3 player, a virtual reality headset, a global positioning system (GPS) or device, a video player, a video camera, a surveillance device or system, a vehicle, a boat, an airship, a virtual machine, a drone, a robot, a handheld communications device, a hospital device, a gaming device or system, an entertainment system, a vehicle computer system, an embedded system controller, a remote control, an appliance, a consumer electronic device, a workstation, an edge device, any combination of these depicted devices, or any other suitable device.

[0097] Exemplary Autonomous Vehicle 11A is a diagram of an example autonomous vehicle 1100 according to some embodiments of the disclosure. Autonomous vehicle 1100 (alternatively referred to herein as “vehicle 1100”) may include, but is not limited to, passenger vehicles, such as cars, trucks, buses, first responder vehicles, shuttles, electric or mopeds, motorcycles, fire engines, police vehicles, ambulances, boats, construction vehicles, submarines, drones, and / or other types of vehicles (e.g., unmanned and / or carrying one or more passengers). Autonomous vehicles are generally described in terms of automation levels as defined by the National Highway Traffic Safety Administration (NHTSA), a division of the U.S. Department of Transportation, and the Society of Automotive Engineers (SAE) "Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicle" (Standard No. J3016-201806 published June 15, 2018, Standard No. J3016-201609 published September 30, 2016, and previous and future versions of this standard). Vehicle 1100 may be capable of functioning according to one or more of levels 3 through 5 of autonomous driving. For example, vehicle 1100 may be capable of conditional automation (Level 3), highly automated (Level 4), and / or fully automated (Level 5), depending on the embodiment.

[0098] The vehicle 1100 may include components such as a vehicle chassis, body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other components. The vehicle 1100 may include a propulsion system 1150, such as an internal combustion engine, a hybrid power plant, a fully electric engine, and / or another propulsion system type. The propulsion system 1150 may be connected to a drive train of the vehicle 1100, which may include a transmission, to enable propulsion of the vehicle 1100. The propulsion system 1150 may be controlled in response to receiving a signal from a throttle / accelerator 1152.

[0099] A steering system 1154, which may include a steering wheel, may be used to steer the vehicle 1100 (e.g., along a desired course or route) when the propulsion system 1150 is operating (e.g., when the vehicle is moving). The steering system 1154 may receive signals from a steering actuator 1156. The steering wheel may be an option for fully automated (Level 5) functionality.

[0100] The brake sensor system 1146 may be used to operate the vehicle brakes in response to receiving a signal from a brake actuator 1148 and / or a brake sensor.

[0101] The controller 1136, which may include one or more system on chip (SoC) 1104 (FIG. 11C) and / or a GPU, can provide signals (e.g., representations of commands) to one or more components and / or systems of the vehicle 1100. For example, the controller can send signals to operate vehicle brakes via one or more brake actuators 1148, operate a steering system 1154 via one or more steering actuators 1156, and operate a propulsion system 1150 via one or more throttle / accelerators 1152. The controller 1136 may include one or more on-board (e.g., integrated) computing devices (e.g., supercomputers) that process sensor signals and output operational commands (e.g., signals representing commands) to enable autonomous driving and / or assist a driver in driving the vehicle 1100. The controllers 1136 may include a first controller 1136 for autonomous driving functions, a second controller 1136 for functional safety functions, a third controller 1136 for artificial intelligence functions (e.g., computer vision), a fourth controller 1136 for infotainment functions, a fifth controller 1136 for redundancy in emergency situations, and / or other controllers. In some instances, a single controller 1136 may handle two or more of the aforementioned functions, and two or more controllers 1136 may handle a single function, and / or any combination thereof.

[0102] The controller 1136 may provide signals to control one or more components and / or systems of the vehicle 1100 in response to sensor data (e.g., sensor inputs) received from one or more sensors. The sensor data may be received from, for example, and without limitation, global navigation satellite system sensors 1158 (e.g., global positioning system sensors), RADAR sensors 1160, ultrasonic sensors 1162, LIDAR sensors 1164, inertial measurement unit (IMU) sensors 1166 (e.g., accelerometers, gyroscopes, magnetic compasses, magnetometers, etc.), microphones 1196, stereo cameras 1168, wide-view cameras 1170 (e.g., fisheye cameras), infrared cameras 1172, surround cameras 1174 (e.g., 360 degree cameras), long-range and / or medium-range cameras 1198, speed sensors 1144 (e.g., for measuring the speed of the vehicle 1100), vibration sensors 1142, steering sensors 1140, brake sensors (e.g., as part of a brake sensor system 1146), and / or other sensor types.

[0103] One or more of the controllers 1136 may receive input (e.g., represented by input data) from the instrument cluster 1132 of the vehicle 1100 and provide output (e.g., represented by output data, display data, etc.) via a human-machine interface (HMI) display 1134, an audible annunciator, a loudspeaker, and / or other components of the vehicle 1100. The output may include information such as vehicle velocity, speed, time, map data (e.g., HD map 1122 of FIG. 11C ), position data (e.g., the position of the vehicle 1100 on a map, etc.), direction, the positions of other vehicles (e.g., occupancy grid), information regarding objects and the status of objects as perceived by the controller 1136, etc. For example, the HMI display 1134 may display information regarding the presence of one or more objects (e.g., road signs, warning signs, changing traffic signals, etc.) and / or a driving maneuver that the vehicle has performed, is performing, or will perform (e.g., changing lanes now, taking exit 34B in 2 miles, etc.).

[0104] Vehicle 1100 further includes a network interface 1124 that can communicate over one or more networks using one or more wireless antennas 1126 and / or a modem. For example, network interface 1124 can be capable of communication over LTE, WCDMA, UMTS, GSM, CDMA2000, etc. Wireless antenna 1126 can also enable communication between objects (e.g., vehicles, mobile devices, etc.) in the environment using local area networks such as Bluetooth, Bluetooth LE, Z-Wave, ZigBee, etc., and / or low power wide-area networks (LPWANs) such as LoRaWAN, SigFox, etc.

[0105] 11B is an illustration of camera positions and fields of view of the example autonomous vehicle 1100 of FIG. 11A, according to some embodiments of the present disclosure. The cameras and their respective fields of view are one example embodiment and are not intended to be limiting. For example, additional and / or alternative cameras may be included and / or the cameras may be located in different positions on the vehicle 1100.

[0106] The camera type of the camera may include, but is not limited to, a digital camera that may be adapted for use with components and / or systems of the vehicle 1100. The camera may be capable of operating at automotive safety integrity level (ASIL) B and / or at another ASIL. The camera type may be capable of any image capture rate, such as 60 frames per second (fps), 120 fps, 240 fps, etc., depending on the embodiment. The camera may be capable of using a rolling shutter, a global shutter, another type of shutter, or a combination thereof. In some instances, the color filter array may include a red clear clear clear (RCCC) color filter array, a red clear clear blue (RCCB) color filter array, a red blue green clear (RBGC) color filter array, a Foveon X3 color filter array, a Bayer sensor (RGGB) color filter array, a monochrome sensor color filter array, and / or another type of color filter array. In some embodiments, clear pixel cameras, such as cameras with RCCC, RCCB, and / or RBGC color filter arrays, may be used in an effort to increase light sensitivity.

[0107] In some instances, one or more of the cameras may be used to perform advanced driver assistance system (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a multi-function mono camera may be installed to provide functions including lane departure warning, traffic sign assist, and intelligent headlamp control. One or more of the cameras (e.g., all of the cameras) may simultaneously record and provide image data (e.g., video).

[0108] One or more of the cameras may be mounted in a mounting part, such as a custom designed (3D printed) part, to filter out stray light and reflections from within the vehicle (e.g., reflections from the dashboard reflected in the windshield mirror) that may interfere with the camera's image data capture capabilities. With reference to a side mirror mounting part, the side mirror part may be custom 3D printed such that the camera mounting plate fits the shape of the side mirror. In some instances, the camera may be integrated into the side mirror. For side view cameras, the cameras may also be integrated into four posts at each corner of the cabin.

[0109] A camera (e.g., a forward-facing camera) with a field of view that includes a portion of the environment in front of the vehicle 1100 may be used for surround view to help identify the forward path and obstacles and, with the aid of one or more controllers 1136 and / or control SoCs, provide information essential to generating an occupancy grid and / or determining a preferred vehicle path. Forward-facing cameras may be used to perform many of the same ADAS functions as LIDAR, including emergency braking, pedestrian detection, and collision avoidance. Forward-facing cameras may also be used for ADAS functions and systems, including other functions such as lane departure warning ("LDW"), autonomous cruise control ("ACC"), and / or traffic sign recognition.

[0110] A variety of cameras may be used in a forward-facing configuration, including, for example, a monocular camera platform that includes a complementary metal oxide semiconductor (CMOS) color imager. Another example may be a wide-view camera 1170 that may be used to understand objects coming into view from the periphery (e.g., pedestrians, crossing traffic, or bicycles). Although only one wide-view camera is shown in FIG. 11B, there may be any number of wide-view cameras 1170 in the vehicle 1100. Additionally, a long-range camera 1198 (e.g., a long-view stereo camera pair) may be used for depth-based object detection, especially for objects for which a neural network has not yet been trained. The long-range camera 1198 may also be used for object detection and classification, as well as basic object tracking.

[0111] One or more stereo cameras 1168 may also be included in the forward-facing configuration. The stereo camera 1168 may include an integrated control unit with an extensible processing unit that may provide programmable logic (FPGA) and a multi-core microprocessor with CAN or Ethernet interfaces integrated on a single chip. Such a unit may be used to generate a 3D map of the vehicle's environment, including distance estimates for all points in the image. An alternative stereo camera 1168 may include a compact stereo vision sensor that may include two camera lenses (one left and one right) and an image processing chip that can measure distances from the vehicle to objects of interest and use the generated information (e.g., metadata) to activate autonomous emergency braking and lane departure warning functions. Other types of stereo cameras 1168 may be used in addition to or instead of those described herein.

[0112] A camera having a field of view that includes a portion of the environment to the side of the vehicle 1100 (e.g., a side-view camera) may be used for surround view, providing information used to create and update the occupancy grid and to generate side impact collision warnings. For example, surround cameras 1174 (e.g., four surround cameras 1174 as shown in FIG. 11B) may be positioned on the vehicle 1100. The surround cameras 1174 may include a wide-view camera 1170, a fish-eye camera, a 360-degree camera, and / or the like. For example, four fish-eye cameras may be positioned at the front, rear, and sides of the vehicle. In an alternative arrangement, the vehicle may use three surround cameras 1174 (e.g., left, right, and rear) and may utilize one or more other cameras (e.g., a forward-facing camera) as a fourth surround view camera.

[0113] A camera having a field of view that includes the portion of the environment behind the vehicle 1100 (e.g., a rear-view camera) may be used for parking assistance, surround view, rear collision warning, and creating and updating an occupancy grid. As described herein, a wide variety of cameras may be used, including, but not limited to, cameras that are also suitable as forward-facing cameras (e.g., long-range and / or mid-range camera 1198, stereo camera 1168, infrared camera 1172, etc.).

[0114] FIG. 11C is a block diagram of an example system architecture of the example autonomous vehicle 1100 of FIG. 11A, according to some embodiments of the present disclosure. It should be understood that this and other arrangements described herein are merely illustrative. Other arrangements and elements (e.g., machines, interfaces, functions, sequences, groupings of functions, etc.) may be used in addition to or instead of those shown, and some elements may be omitted altogether. Furthermore, many of the elements described herein are functional entities that may be implemented as separate or distributed components or in conjunction with other components, and in any suitable combination and location. Various functions described herein as being performed by entities may be implemented by hardware, firmware, and / or software. For example, various functions may be implemented by a processor executing instructions stored in a memory.

[0115] Each of the components, features, and systems of the vehicle 1100 of FIG. 11C are illustrated as being connected via a bus 1102. The bus 1102 may include a controller area network (CAN) data interface (alternatively referred to as a “CAN bus”). The CAN may be a network within the vehicle 1100 that is used to help control various features and functions of the vehicle 1100, such as actuation of brakes, acceleration, braking, steering, windshield wipers, etc. The CAN bus may be configured to have dozens or even hundreds of nodes, each with its own unique identifier (e.g., CAN ID). The CAN bus may be read to find steering wheel angle, ground speed, engine revolutions per minute (RPM), button positions, and / or other vehicle status indicators. The CAN bus may be ASIL B compliant.

[0116] The bus 1102 is described herein as being a CAN bus, but this is not intended to be limiting. For example, FlexRay and / or Ethernet may be used in addition to or as an alternative to a CAN bus. Additionally, a single line is used to represent the bus 1102, but this is not intended to be limiting. There may be any number of buses 1102, which may include, for example, one or more CAN buses, one or more FlexRay buses, one or more Ethernet buses, and / or one or more other types of buses using different protocols. In some instances, two or more buses 1102 may be used to perform different functions and / or may be used for redundancy. For example, a first bus 1102 may be used for collision avoidance functions and a second bus 1102 may be used for operational control. In any instance, each bus 1102 may communicate with any of the components of the vehicle 1100, and two or more buses 1102 may communicate with the same component. In some examples, each SoC 1104, each controller 1136, and / or each computer in the vehicle may have access to the same input data (e.g., input from sensors in the vehicle 1100) and may be connected to a common bus, such as a CAN bus.

[0117] 11A. The controller 1136 may be used for a variety of functions. The controller 1136 may be coupled to any of a variety of other components and systems of the vehicle 1100 and may be used for control of the vehicle 1100, artificial intelligence of the vehicle 1100, infotainment for the vehicle 1100, and / or the like.

[0118] The vehicle 1100 may include a system on a chip (SoC) 1104. The SoC 1104 may include a CPU 1106, a GPU 1108, a processor 1110, a cache 1112, an accelerator 1114, a data store 1116, and / or other components and features not shown. The SoC 1104 may be used to control the vehicle 1100 in a variety of platforms and systems. For example, the SoC 1104 may be coupled in a system (e.g., the system of the vehicle 1100) having an HD map 1122 that can obtain map refreshes and / or updates via a network interface 1124 from one or more servers (e.g., server 1178 of FIG. 11D ).

[0119] The CPU 1106 may include a CPU cluster or CPU complex (alternatively referred to as a "CCPLEX"). The CPU 1106 may include multiple cores and / or L2 caches. For example, in some embodiments, the CPU 1106 may include eight cores in a coherent multiprocessor configuration. In some embodiments, the CPU 1106 may include four dual-core clusters, each with its own dedicated L2 cache (e.g., a 2M B L2 cache). The CPU 1106 (e.g., a CCPLEX) may be configured to support concurrent cluster operation allowing any combination of the CPU 1106 clusters to be active at any given time.

[0120] The CPU 1106 may implement power management capabilities including one or more of the following features: individual hardware blocks may be automatically clock gated when idle to conserve dynamic power, each core clock may be gated when the core is not actively executing instructions by execution of WFI / WFE instructions, each core may be independently power gated, each core cluster may be independently clock gated when all cores are clock gated or power gated, and / or each core cluster may be independently power gated when all cores are power gated. The CPU 1106 may further implement an enhanced algorithm for managing power states where allowed power states and expected wake-up times are specified and the hardware / microcode determines the best power state for entering the cores, clusters, and CCPLEXes. The processing cores may support simplified power state entry sequences in software where work is offloaded to the microcode.

[0121] The GPU 1108 may include an integrated GPU (alternatively referred to herein as an "iGPU"). The GPU 1108 may be programmable and efficient for parallel workloads. In some instances, the GPU 1108 may use an enhanced tensor instruction set. The GPU 1108 may include one or more streaming microprocessors, where each streaming microprocessor may include an L1 cache (e.g., an L1 cache having at least 96 KB storage capacity) and two or more of the streaming microprocessors may share a cache (e.g., an L2 cache having 512 KB storage capacity). In some embodiments, the GPU 1108 may include at least eight streaming microprocessors. The GPU 1108 may use a computational application programming interface (API). In addition, the GPU 1108 may use one or more parallel computing platforms and / or programming models (e.g., NVIDIA's CUDA).

[0122] The GPU 1108 may be power optimized for best performance in automotive and embedded use cases. For example, the GPU 1108 may be fabricated on FinFETs (Fin field-effect transistors). However, this is not intended to be limiting and the GPU 1108 may be fabricated using other semiconductor fabrication processes. Each streaming microprocessor may incorporate several mixed-precision processing cores partitioned into multiple blocks. For example, and without limitation, 64 PF32 cores and 32 PF64 cores may be partitioned into four processing blocks. In such an example, each processing block may be assigned 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two mixed-precision NVIDIA tensor cores for deep learning matrix operations, an L0 instruction cache, a warp scheduler, a dispatch unit, and / or a 64KB register file. In addition, the streaming microprocessor may include independent parallel integer and floating-point data paths to provide efficient execution of workloads having a mix of computational and addressing operations. Streaming microprocessors may include independent thread scheduling capabilities to allow finer grain synchronization and coordination among concurrent threads. Streaming microprocessors may include a combined L1 data cache and shared memory unit to improve performance while simplifying programming.

[0123] The GPU 1108 may include high bandwidth memory (HBM) and / or a 16GB HBM2 memory subsystem to provide about 900GB / sec peak memory bandwidth in some instances. In some instances, synchronous graphics random-access memory (SGRAM), such as graphics double data rate type five synchronous random-access memory (GDDR5), may be used in addition to or in place of the HBM memory.

[0124] The GPU 1108 may include unified memory technology including access counters to enable more accurate movement of memory pages to the processors that access them most frequently, thereby improving the efficiency of storage ranges shared between processors. In some instances, address translation service (ATS) support may be used to enable the GPU 1108 to directly access the CPU 1106 page tables. In such instances, when the GPU 1108 memory management unit (MMU) experiences a miss, an address translation request may be sent to the CPU 1106. In response, the CPU 1106 may consult its page tables for a virtual-to-real mapping of the address and send the translation back to the GPU 1108. As such, the unified memory technology may enable a single unified virtual address space for both the CPU 1106 and GPU 1108 memories, thereby simplifying GPU 1108 programming and porting of applications to the GPU 1108.

[0125] In addition, the GPU 1108 may include access counters that can record the frequency of accesses of the GPU 1108 to the memory of other processors. The access counters can help ensure that memory pages are moved to the physical memory of the processor that is accessing the page most frequently.

[0126] The SoC 1104 may include any number of caches 1112, including those described herein. For example, the cache 1112 may include an L3 cache available to both the CPU 1106 and the GPU 1108 (e.g., connected to both the CPU 1106 and the GPU 1108). The cache 1112 may include a write-back cache that can record line state, such as by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). The L3 cache may include 4 MB or more, depending on the implementation, although smaller cache sizes may be used.

[0127] The SoC 1104 may include an arithmetic logic unit (ALU) that may be utilized in performing processing for any of the various tasks or operations (e.g., processing DNNs) of the vehicle 1100. In addition, the SoC 1104 may include a floating point unit (FPU) (or other math co-processor or numeric co-processor type) for performing mathematical operations within the system. For example, the SoC 1104 may include one or more FPUs integrated as execution units within the CPU 1106 and / or GPU 1108.

[0128] The SoC 1104 may include one or more accelerators 1114 (e.g., hardware accelerators, software accelerators, or a combination thereof). For example, the SoC 1104 may include a hardware acceleration cluster, which may include optimized hardware accelerators and / or large on-chip memory. The large on-chip memory (e.g., 4MB of SRAM) may enable the hardware acceleration cluster to accelerate neural networks and other operations. The hardware acceleration cluster may be used to complement the GPU 1108 and to offload some of the tasks of the GPU 1108 (e.g., to free up more cycles for the GPU 1108 to perform other tasks). As an example, the accelerator 1114 may be used for target workloads that are stable enough to be suitable for acceleration (e.g., perception, convolutional neural networks (CNNs), etc.). As used herein, the term "CNN" may include all types of CNNs, including region-based or regional convolutional neural networks (RCNNs) and Faster RCNNs (e.g., as used for object detection).

[0129] The accelerator 1114 (e.g., a hardware acceleration cluster) may include a deep learning accelerator (DLA). The DLA may include one or more tensor processing units (TPUs) that can be configured to provide an additional 10 trillion operations per second for deep learning applications and inference. The TPU may be an accelerator configured and optimized to perform image processing functions (e.g., CNN, RCNN, etc.). The DLA may further be optimized for a specific set of neural network types and floating point operations, as well as inference. The design of the DLA may provide more performance per millimeter than a general-purpose GPU, greatly exceeding the performance of a CPU. The TPU may perform several functions, including, for example, single instance convolution functions, supporting INT8, INT16, and FP16 data types for both features and weights, as well as post-processor functions.

[0130] The DLA can quickly and efficiently run neural networks, in particular CNNs, on processed or unprocessed data for any of a variety of functions, including but not limited to: CNNs for object identification and detection using data from camera sensors, CNNs for distance estimation using data from camera sensors, CNNs for emergency vehicle detection and identification using data from microphones, CNNs for face recognition and vehicle owner identification using data from camera sensors, and / or CNNs for security and / or safety related events.

[0131] The DLA can perform any function of the GPU 1108, and by using an inference accelerator, for example, a designer can target either the DLA or the GPU 1108 for any function. For example, a designer can focus on processing CNNs and floating point operations on the DLA and offload other functions to the GPU 1108 and / or other accelerators 1114.

[0132] The accelerator 1114 (e.g., hardware acceleration cluster) may include a programmable vision accelerator (PVA), which may alternatively be referred to herein as a computer vision accelerator. The PVA may be designed and configured to accelerate computer vision algorithms for advanced driver assistance systems (ADAS), autonomous driving, and / or augmented reality (AR) and / or virtual reality (VR) applications. The PVA may provide a balance between performance and flexibility. For example, each PVA may include, but is not limited to, any number of reduced instruction set computer (RISC) cores, direct memory access (DMA), and / or any number of vector processors, for example.

[0133] The RISC cores may interact with an image sensor (e.g., an image sensor of any of the cameras described herein), an image signal processor, and / or the like. Each RISC core may include any amount of memory. The RISC cores may use any of a number of protocols, depending on the embodiment. In some instances, the RISC cores may run a real-time operating system (RTOS). The RISC cores may be implemented using one or more integrated circuit devices, application specific integrated circuits (ASICs), and / or memory devices. For example, the RISC cores may include an instruction cache and / or tightly coupled RAM.

[0134] The DMA may allow components of the PVA to access system memory independent of the CPU 1106. The DMA may support any number of features used to provide optimizations to the PVA, including but not limited to supporting multi-dimensional addressing and / or circular addressing. In some instances, the DMA may support up to six or more dimensions of addressing, which may include block width, block height, block depth, horizontal block stepping, vertical block stepping, and / or depth stepping.

[0135] The vector processor may be a programmable processor that may be designed to efficiently and flexibly execute computer vision algorithm programming and provide signal processing capabilities. In some instances, the PVA may include a PVA core and two vector processing subsystem partitions. The PVA core may include a processor subsystem, a DMA engine (e.g., two DMA engines), and / or other peripherals. The vector processing subsystem may act as the primary processing engine of the PVA and may include a vector processing unit (VPU), an instruction cache, and / or a vector memory (e.g., VMEM). The VPU core may include a digital signal processor, such as, for example, a single instruction, multiple data (SIMD), very long instruction word (VLIW) digital signal processor. The combination of SIMD and VLIW may increase throughput and speed.

[0136] Each vector processor may include an instruction cache and may be coupled to dedicated memory. As a result, in some instances, each vector processor may be configured to execute independently of other vector processors. In other instances, the vector processors included in a particular PVA may be configured to employ data parallelism. For example, in some embodiments, multiple vector processors included in a single PVA may execute the same computer vision algorithm, but on different regions of an image. In other instances, the vector processors included in a particular PVA may execute different computer vision algorithms simultaneously on the same image, or even execute different algorithms on sequential images or portions of an image. In particular, any number of PVAs may be included in a hardware accelerated cluster, and any number of vector processors may be included in each PVA. In addition, the PVAs may include additional error correcting code (ECC) memory to increase overall system security.

[0137] The accelerator 1114 (e.g., a hardware acceleration cluster) may include a computer vision network-on-chip and SRAM to provide high bandwidth, low latency SRAM for the accelerator 1114. In some instances, the on-chip memory may include, for example and without limitation, at least 4 MB of SRAM consisting of eight field configurable memory blocks that may be accessible by both the PVA and DLA. Each pair of memory blocks may include an advanced peripheral bus (APB) interface, configuration circuitry, controllers, and multiplexers. Any type of memory may be used. The PVA and DLA may access the memory through a backbone that provides the PVA and DLA with high speed access to the memory. The backbone may include a computer vision network-on-chip that interconnects the PVA and DLA to the memory (e.g., using the APB).

[0138] The computer vision network-on-chip may include an interface that determines, prior to transmission of any control signals / addresses / data, that both the PVA and DLA provide ready and valid signals. Such an interface may provide separate phases and separate channels for transmitting control signals / addresses / data, as well as burst type communication for continuous data transfer. This type of interface may follow ISO26262 or IEC61508 standards, although other standards and protocols may be used.

[0139] In some examples, the SoC 1104 may include a real-time ray tracing hardware accelerator, such as described in U.S. Patent Application No. 16 / 101,232, filed August 10, 2018. The real-time ray tracing hardware accelerator may be used to quickly and efficiently determine the location and scale of objects (e.g., within a world model) to generate real-time visualization simulations for RADAR signal interpretation, for acoustic propagation synthesis and / or analysis, for simulation of SONAR systems, for general wave propagation simulation, for comparison against LIDAR data for localization and / or other functions, and / or for other uses. In some examples, one or more tree traversal units (TTUs) may be used to perform one or more ray tracing related operations.

[0140] The accelerator 1114 (e.g., a hardware accelerator cluster) has a variety of applications for autonomous driving. The PVA may be a programmable vision accelerator that can be used for critical processing stages in ADAS and autonomous vehicles. The capabilities of the PVA are well suited to the domain of algorithms that require predictable processing at low power and low latency. In other words, the PVA works well with semi-dense or dense regular computations on small data sets that require predictable execution times with low latency and low power. Hence, in the context of a platform for autonomous vehicles, the PVA is designed to run classical computer vision algorithms, since the PVA is efficient in object detection and operating on integer computations.

[0141] For example, according to one embodiment of the present technology, the PVA is used to perform computer stereo vision. A semi-global matching based algorithm may be used in some instances, but this is not intended to be limiting. Many applications for Level 3-5 autonomous driving require motion estimation / stereo matching on the fly (e.g., structure from motion, pedestrian recognition, lane detection, etc.). The PVA can perform computer stereo vision functions with input from two monocular cameras.

[0142] In some instances, PVA may be used to perform dense optical flow by processing raw RADAR data (e.g., using a 4D Fast Fourier Transform) to provide processed RADAR. In other instances, PVA is used for time of flight depth processing, for example, by processing raw time of flight data to provide processed time of flight data.

[0143] DLA can be used to implement any type of network to enhance control and driving safety, including, for example, a neural network that outputs a measure of the confidence of each object detection. Such confidence value can be interpreted as a probability or as providing a relative "weight" of each detection compared to other detections. This confidence value allows the system to make further decisions regarding which detections should be considered true positive detections rather than false positive detections. For example, the system can set a confidence threshold and consider only detections above the threshold as true positive detections. In an automatic emergency braking (AEB) system, a false positive detection would cause the vehicle to automatically perform emergency braking, which is clearly undesirable. Therefore, only the most confident detections should be considered as triggers for AEB. DLA can implement a neural network that regresses the confidence values. The neural network may receive as its inputs at least some subset of parameters, such as bounding box dimensions, ground plane estimates obtained (e.g., from another subsystem), inertial measurement unit (IMU) sensor 1166 outputs that correlate with the vehicle 1100 orientation, distance, and 3D position estimates of objects obtained from the neural network and / or other sensors (e.g., LIDAR sensor 1164 or RADAR sensor 1160), and so forth.

[0144] The SoC 1104 may include a data store 1116 (e.g., a memory). The data store 1116 may be an on-chip memory of the SoC 1104 and may store the neural network to be executed on the GPU and / or DLA. In some instances, the data store 1116 may have a capacity large enough to store multiple instances of the neural network for redundancy and safety. The data store 1112 may comprise an L2 or L3 cache 1112. References to the data store 1116 may include references to memory associated with the PVA, DLA, and / or other accelerators 1114, as described herein.

[0145] The SoC 1104 may include one or more processors 1110 (e.g., embedded processors). The processors 1110 may include a boot and power management processor, which may be a dedicated processor and subsystem for handling boot power and management capabilities and associated security enforcement. The boot and power management processor may be part of the SoC 1104 boot sequence and may provide run-time power management services. The boot power and management processor may provide clock and voltage programming, assist with system low power state transitions, management of the SoC 1104 thermal and temperature sensors, and / or management of the SoC 1104 power states. Each temperature sensor may be implemented as a ring oscillator whose output frequency is proportional to temperature, and the SoC 1104 may use the ring oscillator to detect the temperature of the CPU 1106, the GPU 1108, and / or the accelerator 1114. If the temperature is determined to exceed a threshold, the boot and power management processor may enter a temperature fault routine, place the SoC 1104 in a lower power state and / or place the vehicle 1100 in a Chauffeur safe shutdown mode (e.g., bring the vehicle 1100 to a safe shutdown).

[0146] The processor 1110 may further include a set of embedded processors that may perform the functions of an audio processing engine. The audio processing engine may be an audio subsystem that allows full hardware support of multi-channel audio through multiple interfaces and a wide and flexible range of audio I / O interfaces. In some instances, the audio processing engine is a dedicated processor core that includes a digital signal processor with dedicated RAM.

[0147] The processor 1110 may further include an always-on processor engine that can provide the necessary hardware features to support low power sensor management and wake use cases. The always-on processor engine may include a processor core, tightly coupled RAM, support peripherals (e.g., timers and interrupt controllers), various I / O controller peripherals, and routing logic.

[0148] The processor 1110 may further include a safety cluster engine that includes a processor subsystem dedicated to handling safety management of automotive applications. The safety cluster engine may include two or more processor cores, tightly coupled RAM, supporting peripherals (e.g., timers, interrupt controllers, etc.), and / or routing logic. In a safety mode, the two or more cores may operate in lockstep mode and function as a single core with comparison logic to detect any differences between their operations.

[0149] The processor 1110 may further include a real-time camera engine, which may include a dedicated processor subsystem for handling real-time camera management.

[0150] The processor 1110 may further include a high dynamic range signal processor, which may include an image signal processor, which is a hardware engine that is part of the camera processing pipeline.

[0151] The processor 1110 may include a video image compositor, which may be a processing block (e.g., implemented in a microprocessor) that implements video post-processing functions required by the video playback application to produce a final image for the player window. The video image compositor may perform lens distortion correction on the wide-view camera 1170, on the surround camera 1174, and / or on the in-cabin surveillance camera sensor. The in-cabin surveillance camera sensor is preferably monitored by a neural network running on another instance of the advanced SoC, configured to identify in-cabin events and respond appropriately. The in-cabin system may perform lip reading to activate cellular service and make phone calls, dictate emails, change the vehicle's destination, activate or change the vehicle's infotainment system and settings, or provide voice-activated web surfing. Certain features are only available to the driver when operating in autonomous mode and are disabled otherwise.

[0152] The video image synthesizer may include enhanced temporal noise reduction for both spatial and temporal noise reduction. For example, when motion occurs in the video, the noise reduction reduces the weight of information provided by adjacent frames and appropriately weights the spatial information. When an image or part of an image does not contain motion, the temporal noise reduction performed by the video image synthesizer may use information from previous images to reduce noise in the current image.

[0153] The video image composer may also be configured to perform stereo rectification on the input stereo lens frames. The video image composer may further be used for user interface compositing when the operating system desktop is in use and the GPU 1108 is not required to continuously render new surfaces. Even when the GPU 1108 is powered on and actively performing 3D rendering, the video image composer may be used to offload the GPU 1108 for improved performance and responsiveness.

[0154] The SoC1104 may further include a mobile industry processor interface (MIPI) camera serial interface for receiving video and input from a camera, a high-speed interface, and / or a video input block that may be used for camera and related pixel input functions. The SoC1104 may further include an input / output controller that may be controlled by software and that may be used to receive I / O signals that are not committed to a specific role.

[0155] The SoC 1104 may further include a wide range of peripheral interfaces to enable peripherals, audio codecs, power management, and / or communication with other devices. The SoC 1104 may be used to process data from cameras (e.g., connected via gigabit multimedia serial links and Ethernet), sensors (e.g., LIDAR sensors 1164, RADAR sensors 1160, etc., which may be connected via Ethernet), data from the bus 1102 (e.g., vehicle 1100 speed, steering wheel position, etc.), and data from the GNSS sensor 1158 (e.g., connected via Ethernet or CAN bus). The SoC 1104 may further include a dedicated high-performance mass storage controller, which may include its own DMA engine and may be used to offload the CPU 1106 from routine data management tasks.

[0156] The SoC 1104 may be an end-to-end platform with a flexible architecture that spans levels 3-5 of automation, thereby providing a comprehensive functional safety architecture that leverages and efficiently uses computer vision and ADAS techniques for diversity and redundancy, and provides a platform for a flexible, reliable driving software stack along with deep learning tools. The SoC 1104 may be faster, more reliable, and more energy- and space-efficient than traditional systems. For example, when the accelerator 1114 is combined with the CPU 1106, GPU 1108, and data store 1106, the SoC 1104 may provide a fast and efficient platform for levels 3-5 of autonomous vehicles.

[0157] Thus, the present technology provides capabilities and functionality that cannot be achieved by conventional systems. For example, computer vision algorithms may be executed on a CPU, which may be configured using a high-level programming language, such as the C programming language, to execute a wide variety of processing algorithms across a wide variety of visual data. However, CPUs often cannot meet the performance requirements of many computer vision applications, such as those related to execution time and power consumption. In particular, many CPUs cannot execute complex object detection algorithms in real time, which is a requirement for in-vehicle ADAS applications and for practical Level 3-5 autonomous vehicles.

[0158] In contrast to conventional systems, by providing a CPU complex, a GPU complex, and a hardware acceleration cluster, the techniques described herein allow multiple neural networks to be run simultaneously and / or sequentially and the results to be combined to enable Level 3-5 autonomous driving capabilities. For example, a CNN running on the DLA or dGPU (e.g., GPU1120) can include text and word recognition that allows the supercomputer to read and understand traffic signs, including signs for which the neural network was not specifically trained. The DLA can further include neural networks that can identify, interpret, and provide a semantic understanding of the signs and pass the semantic understanding to a route planning module running on the CPU complex.

[0159] As another example, multiple neural networks may be run simultaneously, as required for level 3, 4, or 5 driving. For example, a warning sign consisting of "Caution: Flashing lights indicate icy conditions" along with a lightning flash may be interpreted by several neural networks, independently or collectively. The sign itself may be identified as a traffic sign by a first deployed neural network (e.g., a trained neural network), and the text "Flashing lights indicate icy conditions" may be interpreted by a second deployed neural network that informs the vehicle's route planning software (preferably running on the CPU complex) that icy conditions exist when a flashing light is detected. The flashing light may be identified by running a third deployed neural network through multiple frames, informing the vehicle's route planning software of the presence (or absence) of the flashing light. All three neural networks may be run simultaneously, such as within the DLA and / or on the GPU 1108.

[0160] In some instances, a CNN for facial recognition and vehicle owner identification can use data from the camera sensors to identify the presence of a legitimate driver and / or owner of the vehicle 1100. An always-on sensor processing engine can be used to unlock the vehicle and turn on the lights when the owner approaches the driver's door, and in security mode, to disable vehicle operation when the owner leaves the vehicle. In this manner, the SoC 1104 provides security against theft and / or carjacking.

[0161] In another example, a CNN for emergency vehicle detection and identification can detect and identify emergency vehicle sirens using data from microphone 1196. In contrast to conventional systems that use general classifiers to detect sirens and manually extract features, SoC 1104 uses CNNs for classification of environmental and urban sounds, as well as classification of visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative terminal velocity of emergency vehicles (e.g., by using the Doppler effect). The CNN can also be trained to identify emergency vehicles specific to the local area in which the vehicle is operating, as identified by the GNSS sensor 1158. Thus, for example, when operating in Europe, the CNN will attempt to detect European sirens, and when in the United States, the CNN will attempt to identify only North American sirens. After an emergency vehicle is detected, a control program can be used to perform emergency vehicle safety routines, such as slowing the vehicle down, stopping at the side of the road, parking the vehicle, and / or idling the vehicle, with the assistance of ultrasonic sensor 1162, until the emergency vehicle has passed.

[0162] The vehicle may include a CPU 1118 (e.g., a separate CPU, or a dCPU) that may be coupled to the SoC 1104 via a high-speed interconnect (e.g., PCIe). The CPU 1118 may include, for example, an X86 processor. The CPU 1118 may be used to perform any of a variety of functions, including, for example, reconciling potentially inconsistent results between the ADAS sensors and the SoC 1104, and / or monitoring the status and health of the controller 1136 and / or the infotainment SoC 1130.

[0163] The vehicle 1100 may include a GPU 1120 (e.g., a discrete GPU, or a dGPU) that may be coupled to the SoC 1104 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The GPU 1120 may provide additional artificial intelligence functionality, such as by running redundant and / or different neural networks, and may be used to train and / or update neural networks based on input (e.g., sensor data) from sensors of the vehicle 1100.

[0164] The vehicle 1100 may further include a network interface 1124, which may include one or more wireless antennas 1126 (e.g., one or more wireless antennas for different communication protocols, such as a cellular antenna, a Bluetooth antenna, etc.). The network interface 1124 may be used to enable wireless connections with the cloud over the Internet (e.g., with the server 1178 and / or other network devices), with other vehicles, and / or with computing devices (e.g., passenger client devices). To communicate with other vehicles, a direct link may be established between the two vehicles and / or an indirect link may be established (e.g., through a network and via the Internet). A direct link may be provided using a vehicle-to-vehicle communication link. The vehicle-to-vehicle communication link may provide the vehicle 1100 information about vehicles in close proximity to the vehicle 1100 (e.g., vehicles in front of, beside, and / or behind the vehicle 1100). This function may be part of a collaborative adaptive cruise control function of the vehicle 1100.

[0165] The network interface 1124 may include a SoC that provides modulation and demodulation functions and enables the controller 1136 to communicate over a wireless network. The network interface 1124 may include a radio frequency front end for upconversion from baseband to radio frequency and downconversion from radio frequency to baseband. The frequency conversion may be performed through well known processes and / or may be performed using a superheterodyne process. In some instances, the radio frequency front end functions may be provided by a separate chip. The network interface may include wireless functions for communicating via LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-Wave, ZigBee, LoRaWAN, and / or other wireless protocols.

[0166] The vehicle 1100 may further include a data store 1128, which may include off-chip (e.g., off-SoC 1104) storage. The data store 1128 may include one or more memory elements, including RAM, SRAM, DRAM, VRAM, flash, hard disk, and / or other components and / or devices capable of storing at least one bit of data.

[0167] The vehicle 1100 may further include a GNSS sensor 1158. The GNSS sensor 1158 (e.g., a GPS, an aided GPS sensor, a differential GPS (DGPS) sensor, etc.) aids in mapping, perception, occupancy grid generation, and / or route planning functions. Any number of GNSS sensors 1158 may be used, including, but not limited to, a GPS using a USB connector with an Ethernet to serial (RS-232) bridge, for example.

[0168] The vehicle 1100 may further include a RADAR sensor 1160. The RADAR sensor 1160 may be used by the vehicle 1100 for long-range vehicle detection, even in darkness and / or severe weather conditions. The RADAR functional safety level may be ASIL B. In some instances, the RADAR sensor 1160 may use the CAN and / or bus 1102 for control and to access object tracking data (e.g., to transmit data generated by the RADAR sensor 1160), with access to Ethernet to access raw data. A wide variety of RADAR sensor types may be used. For example, and without limitation, the RADAR sensor 1160 may be suitable for front, rear, and side RADAR use. In some instances, a pulsed Doppler RADAR sensor is used.

[0169] The RADAR sensor 1160 may include different configurations, such as long range with a narrow field of view, short range with a wide field of view, and short range side coverage. In some instances, the long range RADAR may be used for adaptive cruise control functions. Long range RADAR systems may provide a wide field of view achieved by two or more independent scans, such as within a 250m range. The RADAR sensor 1160 may help distinguish between static and moving objects and may be used by ADAS systems for emergency brake assist and forward collision warning. Long range RADAR sensors may include monostatic multimodal RADAR with multiple (e.g., six or more) fixed RADAR antennas and high speed CAN and FlexRay interfaces. In one instance with six antennas, the center four antennas may create a focused beam pattern designed to record the surroundings of the vehicle 1100 at high speeds with minimal interference from traffic in adjacent lanes. The other two antennas may widen the field of view, allowing for quick detection of vehicles entering or leaving the lane of the vehicle 1100.

[0170] As an example, a medium-range RADAR system may include a range of up to 1160m (front) or 80m (rear) and a field of view of up to 42 degrees (front) or 1150 degrees (rear). A short-range RADAR system may include, but is not limited to, a RADAR sensor designed to be mounted on either end of the rear bumper. When mounted on either end of the rear bumper, such a RADAR sensor system can create two beams that constantly monitor the blind spots behind and adjacent to the vehicle.

[0171] Short-range RADAR systems may be used in ADAS systems for blind spot detection and / or lane change assist.

[0172] The vehicle 1100 may further include ultrasonic sensors 1162. The ultrasonic sensors 1162, which may be located on the front, rear, and / or sides of the vehicle 1100, may be used for parking assist and / or for creating and updating an occupancy grid. A wide variety of ultrasonic sensors 1162 may be used, with different ultrasonic sensors 1162 being used for different ranges of detection (e.g., 2.5m, 4m). The ultrasonic sensors 1162 may operate at a functional safety level of ASIL B.

[0173] The vehicle 1100 may include a LIDAR sensor 1164. The LIDAR sensor 1164 may be used for object and pedestrian detection, emergency braking, collision avoidance, and / or other functions. The LIDAR sensor 1164 may be functional safety level ASIL B. In some instances, the vehicle 1100 may include multiple (e.g., 2, 4, 6, etc.) LIDAR sensors 1164 that may use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).

[0174] In some instances, the LIDAR sensor 1164 may have the ability to provide a list of objects and their distances in a 360-degree field of view. A commercially available LIDAR sensor 1164 may have an advertised range of about 1100 m, for example, with an accuracy of 2 cm to 3 cm, and with support for an 1100 Mbps Ethernet connection. In some instances, one or more non-protruding LIDAR sensors 1164 may be used. In such instances, the LIDAR sensor 1164 may be implemented as a small device that may be integrated into the front, rear, sides, and / or corners of the vehicle 1100. In such instances, the LIDAR sensor 1164 may have a range of 200 m even for low-reflecting objects, and provide up to 120 degrees horizontal and 35 degrees vertical field of view. A front-mounted LIDAR sensor 1164 may be configured for a horizontal field of view between 45 degrees and 135 degrees.

[0175] In some instances, LIDAR technology such as 3D flash LIDAR may also be used. 3D flash LIDAR uses a laser flash as a transmitter to illuminate the surroundings of the vehicle up to about 200 m. The flash LIDAR unit includes a receptor that records the laser pulse transit time and the reflected light on each pixel, which in turn corresponds to the range from the vehicle to the object. Flash LIDAR may allow a highly accurate and distortion-free image of the surroundings to be generated with every laser flash. In some instances, four flash LIDAR sensors may be deployed, one on each side of the vehicle 1100. Available 3D flash LIDAR systems include solid-state 3D steering array LIDAR cameras with no moving parts other than the blower (e.g., non-scanning LIDAR devices). Flash LIDAR devices may use 5 nanosecond Class I (eye-safe) laser pulses per frame and may capture reflected laser light in the form of a 3D range point cloud and coregistered intensity data. By using flash LIDAR, and because flash LIDAR is a solid-state device with no moving parts, the LIDAR sensor 1164 may be less susceptible to motion blur, vibration, and / or shock.

[0176] The vehicle may further include an IMU sensor 1166. In some instances, the IMU sensor 1166 may be positioned at the center of the rear axle of the vehicle 1100. The IMU sensor 1166 may include, for example, but not limited to, an accelerometer, a magnetometer, a gyroscope, a magnetic compass, and / or other sensor types. In some instances, such as in a 6-axis application, the IMU sensor 1166 may include an accelerometer and a gyroscope, while in a 9-axis application, the IMU sensor 1166 may include an accelerometer, a gyroscope, and a magnetometer.

[0177] In some embodiments, the IMU sensor 1166 may be implemented as a miniature, high-performance GPS-Aided Inertial Navigation System (GPS / INS) that combines micro-electro-mechanical system (MEMS) inertial sensors, highly sensitive GPS receivers, and advanced Kalman filtering algorithms to provide estimates of position, velocity, and attitude. As such, in some instances, the IMU sensor 1166 may enable the vehicle 1100 to estimate heading without requiring input from a magnetic sensor by directly observing and correlating changes in velocity from the GPS to the IMU sensor 1166. In some instances, the IMU sensor 1166 and the GNSS sensor 1158 may be combined in a single integrated unit.

[0178] The vehicle may include microphones 1196 placed in and / or around the vehicle 1100. The microphones 1196 may be used for emergency vehicle detection and identification, among other things.

[0179] The vehicle may further include any number of camera types, including stereo cameras 1168, wide view cameras 1170, infrared cameras 1172, surround cameras 1174, long and / or mid-range cameras 1198, and / or other camera types. The cameras may be used to capture image data around the entire exterior of the vehicle 1100. The type of cameras used depends on the embodiment and requirements of the vehicle 1100, and any combination of camera types may be used to achieve the required coverage around the vehicle 1100. Additionally, the number of cameras may vary depending on the embodiment. For example, the vehicle may include six cameras, seven cameras, ten cameras, twelve cameras, and / or another number of cameras. The cameras may support, by way of example only, Gigabit Multimedia Serial Link (GMSL) and / or Gigabit Ethernet, without limitation. Each camera is described in further detail herein with respect to FIGS. 11A and 11B.

[0180] The vehicle 1100 may further include a vibration sensor 1142. The vibration sensor 1142 may measure vibrations of vehicle components, such as an axle. For example, a change in vibration may indicate a change in the surface of the road. In another example, when two or more vibration sensors 1142 are used, the difference in vibration may be used to determine friction or slippage of the road surface (e.g., when the difference in vibration is between a powered axle and a free-running axle).

[0181] The vehicle 1100 may include an ADAS system 1138. In some instances, the ADAS system 1138 may include a SoC. The ADAS system 1138 may include autonomous / adaptive / automatic cruise control (ACC), cooperative adaptive cruise control (CACC), forward crash warning (FCW), automatic emergency braking (AEB), lane departure warning (LDW), lane keep assist (LKA), blind spot warning (BSW), rear cross-traffic warning (RCTW), collision warning system (CWS), lane centering (LC), and / or other features and functions.

[0182] The ACC system may use a RADAR sensor 1160, a LIDAR sensor 1164, and / or a camera. The ACC system may include longitudinal ACC and / or lateral ACC. Longitudinal ACC monitors and controls the distance to the vehicle directly ahead of the vehicle 1100 and automatically adjusts the vehicle speed to maintain a safe distance from the vehicle ahead. Lateral ACC performs distance keeping and advises the vehicle 1100 to change lanes when necessary. Lateral ACC is related to other ADAS applications such as LCA and CWS.

[0183] CACC uses information from other vehicles, which may be received from other vehicles via a wireless link via the network interface 1124 and / or the wireless antenna 1126, or indirectly via a network connection (e.g., via the Internet). A direct link may be provided by a vehicle-to-vehicle (V2V) communication link, while an indirect link may be an infrastructure-to-vehicle (I2V) communication link. In general, a V2V communication concept provides information about the immediately preceding vehicle (e.g., the vehicle directly ahead of the vehicle 1100, in the same lane as the vehicle 1100), while an I2V communication concept provides information about the traffic further ahead. A CACC system may include either or both I2V and V2V information sources. Given the information of the vehicles ahead of the vehicle 1100, CACC may be more reliable and may have the potential to make traffic flow smoother and reduce congestion on the roads.

[0184] The FCW system is designed to warn the driver of hazards so that the driver can take corrective action. The FCW system uses a forward-facing camera and / or RADAR sensor 1160 coupled to a dedicated processor, DSP, FPGA, and / or ASIC, electrically coupled to driver feedback such as a display, speaker, and / or vibration components. The FCW system can provide warnings in the form of audio, visual alerts, vibrations, and / or quick brake pulses.

[0185] An AEB system can detect an imminent forward collision with another vehicle or other object and automatically apply the brakes if the driver does not take corrective action within specified time or distance parameters. The AEB system can use a forward-facing camera and / or RADAR sensor 1160 coupled to a dedicated processor, DSP, FPGA, and / or ASIC. When the AEB system detects a hazard, the AEB system typically first alerts the driver to take corrective action to avoid the collision, and if the driver does not take corrective action, the AEB system can automatically apply the brakes as part of an effort to prevent, or at least mitigate, the effects of the predicted collision. The AEB system can include techniques such as Dynamic Brake Support and / or Collision Imminent Braking.

[0186] The LDW system provides visual, audible, and / or tactile warnings, such as vibrations in the steering wheel or seat, to alert the driver when the vehicle 1100 crosses a lane marking. The LDW system does not activate when the driver indicates an intentional lane departure by activating a turn signal. The LDW system may use a forward-facing camera coupled to a dedicated processor, DSP, FPGA, and / or ASIC electrically coupled to driver feedback, such as a display, speaker, and / or vibration components.

[0187] The LKA system is a modification of the LDW system, which provides steering input or braking to correct the vehicle 1100 if it begins to drift out of its lane.

[0188] BSW systems detect and warn vehicle drivers in the vehicle's blind spots. BSW systems can provide visual, audible, and / or tactile warnings to indicate that merging or changing lanes is unsafe. The system can provide additional warnings when the driver uses a turn signal. BSW systems can use rear-facing cameras and / or RADAR sensors coupled to dedicated processors, DSPs, FPGAs, and / or ASICs electrically coupled to driver feedback, e.g., displays, speakers, and / or vibration components.

[0189] The RCTW system can provide visual, audible, and / or tactile notifications when an object is detected outside the range of the rear camera when the vehicle 1100 is backing up. Some RCTW systems include AEB to ensure vehicle brakes are applied to avoid a collision. The RCTW system can use one or more rear-facing RADAR sensors 1160 coupled to a dedicated processor, DSP, FPGA, and / or ASIC electrically coupled to driver feedback, e.g., a display, speaker, and / or vibration components.

[0190] Because conventional ADAS systems alert the driver and allow the driver to determine whether a safety condition truly exists and act accordingly, conventional ADAS systems can be prone to producing false positive results that are usually not catastrophic but can be annoying and distracting to the driver. However, in an autonomous vehicle 1100, the vehicle 1100 itself must decide whether to listen to the results from the primary computer or the secondary computer (e.g., the first controller 1136 or the second controller 1136) when the results are conflicting. For example, in some embodiments, the ADAS system 1138 may be a backup and / or secondary computer to provide perception information to a backup computer rationality module. The backup computer rationality monitor can run redundant and diverse software on hardware components to detect failures in perception and dynamic driving tasks. Output from the ADAS system 1138 may be provided to a supervisory MCU. When the outputs from the primary and secondary computers conflict, the supervisory MCU must decide how to reconcile the conflict to ensure safe operation.

[0191] In some instances, the primary computer may be configured to provide a reliability score to the overseer MCU indicating the reliability of the primary computer in a selected outcome. If the reliability score exceeds a threshold, the overseer MCU may follow the instructions of the primary computer regardless of whether the secondary computers provide conflicting or inconsistent results. If the reliability score does not meet the threshold, and if the primary and secondary computers show different (e.g., conflicting) results, the overseer MCU may arbitrate between the computers to determine the appropriate outcome.

[0192] The supervisory MCU may be configured to execute a neural network that is trained and configured to determine conditions under which the secondary computer provides a false alarm based on the outputs from the primary and secondary computers. Thus, the neural network in the supervisory MCU may learn when the output of the secondary computer may be trusted and when it may not be trusted. For example, when the secondary computer is a RADAR-based FCW system, the neural network in the supervisory MCU may learn when the FCW identifies a metal object that is not actually a hazard, such as a sewer grate or manhole cover, that triggers an alarm. Similarly, when the secondary computer is a camera-based LDW system, the neural network in the supervisory MCU may learn to ignore the LDW when a bicyclist or pedestrian is present and lane departure is, in fact, the safest maneuver. In embodiments that include a neural network executing on the supervisory MCU, the supervisory MCU may include at least one of a DLA or a GPU suitable for executing the neural network with associated memory. In a preferred embodiment, the director MCU may comprise and / or be included as a component of the SoC 1104.

[0193] In other instances, the ADAS system 1138 may include a secondary computer that performs ADAS functions using classical rules of computer vision. As such, the secondary computer may use classical computer vision rules (if-then) and the presence of a neural network in the supervisory MCU may improve reliability, safety and performance. For example, the diverse implementations and intentional non-identity may make the overall system more fault-tolerant, especially to failures caused by software (or software-hardware interface) functions. For example, if there is a software bug or error in the software running on the primary computer and non-identical software code running on the secondary computer provides the same overall result, the supervisory MCU may have greater confidence that the overall result is correct and that a bug in the software or hardware on the primary computer has not caused a critical error.

[0194] In some instances, the output of the ADAS system 1138 may be fed to a perception block of the primary computer and / or a dynamic driving task block of the primary computer. For example, if the ADAS system 1138 indicates a forward collision warning due to an object directly ahead, the perception block may use this information when identifying the object. In other instances, the secondary computer may have its own neural network that is trained as described herein, thus reducing the risk of false positives.

[0195] The vehicle 1100 may further include an infotainment SoC 1130 (e.g., an in-vehicle infotainment system (IVI)). Although shown and described as an SoC, the infotainment system may not be an SoC and may include two or more separate components. The infotainment SoC 1130 may include a combination of hardware and software that may be used to provide audio (e.g., music, personal digital assistant, navigation instructions, news, radio, etc.), video (e.g., TV, movies, streaming, etc.), telephony (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and / or information services (e.g., navigation system, rear park assist, wireless data system, vehicle related information such as fuel level, total distance traveled, brake fuel level, oil level, door open / close, air filter information, etc.) to the vehicle 1100. For example, the infotainment SoC 1130 may be a radio, a disc player, a navigation system, a video player, USB and Bluetooth connectivity, a car computer, in-car entertainment, Wi-Fi, steering wheel audio controls, hands-free voice control, a heads-up display (HUD), an HMI display 1134, a telematics device, a control panel (e.g., for controlling and / or interacting with various components, features, and / or systems), and / or other components. The infotainment SoC 1130 may further be used to provide information (e.g., visual and / or audible) to a user of the vehicle, such as information from an ADAS system 1138, autonomous driving information such as planned vehicle maneuvers, trajectory, surrounding environment information (e.g., intersection information, vehicle information, road information, etc.), and / or other information.

[0196] The infotainment SoC 1130 may include GPU functionality. The infotainment SoC 1130 may communicate with other devices, systems, and / or components of the vehicle 1100 via the bus 1102 (e.g., CAN bus, Ethernet, etc.). In some instances, the infotainment SoC 1130 may be coupled to a supervisory MCU such that the infotainment system's GPU can perform some self-drive functions in the event that the primary controller 1136 (e.g., the vehicle's 1100 primary and / or backup computer) fails. In such instances, the infotainment SoC 1130 may place the vehicle 1100 in a Chauffeur safe shutdown mode, as described herein.

[0197] The vehicle 1100 may further include an instrument cluster 1132 (e.g., a digital dash, an electronic instrument cluster, a digital instrument panel, etc.). The instrument cluster 1132 may include a controller and / or a supercomputer (e.g., a separate controller or a supercomputer). The instrument cluster 1132 may include a set of instruments such as a speedometer, a fuel level, an oil pressure, a tachometer, an odometer, a turn signal, a gear shift position indicator, a seat belt warning light, a parking brake warning light, an engine malfunction light, an airbag (SRS) system information, lighting controls, safety system controls, navigation information, etc. In some instances, information may be displayed and / or shared between the infotainment SoC 1130 and the instrument cluster 1132. In other words, the instrument cluster 1132 may be included as part of the infotainment SoC 1130, or vice versa.

[0198] 11D is a system diagram of communication between a cloud-based server and an example autonomous vehicle 1100 of FIG. 11A in accordance with some embodiments of the present disclosure. The system 1176 may include a server 1178, a network 1190, and a vehicle including the vehicle 1100. The server 1178 may include multiple GPUs 1184(A)-1184(H) (collectively referred to herein as GPUs 1184), PCIe switches 1182(A)-1182(H) (collectively referred to herein as PCIe switches 1182), and / or CPUs 1180(A)-1180(B) (collectively referred to herein as CPUs 1180). The GPUs 1184, CPUs 1180, and PCIe switches may be interconnected with a high-speed interconnect, such as, but not limited to, an NVLink interface 1188 developed by NVIDIA and / or a PCIe connection 1186. In some instances, the GPUs 1184 are connected via an NVLink and / or NVSwitch SoC, and the GPUs 1184 and PCIe switch 1182 are connected via a PCIe interconnect. Although eight GPUs 1184, two CPUs 1180, and two PCIe switches are illustrated, this is not intended to be limiting. Depending on the embodiment, each server 1178 may include any number of GPUs 1184, CPUs 1180, and / or PCIe switches. For example, the servers 1178 may each include 8, 16, 32, and / or more GPUs 1184.

[0199] The server 1178 can receive image data from the vehicle over the network 1190, the image data representing images showing unexpected or changed road conditions, such as recently started road construction. The server 1178 can transmit the neural network 1192, the updated neural network 1192, and / or the map information 1194, including information about traffic and road conditions, to the vehicle over the network 1190. The updates to the map information 1194 can include updates to the HD map 1122, such as information about construction sites, potholes, detours, flooding, and / or other obstacles. In some instances, the neural network 1192, the updated neural network 1192, and / or the map information 1194 may have arisen from new training and / or experience represented in data received from any number of vehicles in the environment and / or based on training performed at a data center (e.g., using the server 1178 and / or other servers).

[0200] The server 1178 may be used to train a machine learning model (e.g., a neural network) based on the training data. The training data may be generated by the vehicle and / or generated in a simulation (e.g., using a game engine). In some instances, the training data is tagged (e.g., if the neural network benefits from supervised learning) and / or undergoes other pre-processing, while in other instances, the training data is not tagged and / or pre-processed (e.g., if the neural network does not require supervised learning). The training may be performed according to any one or more classes of machine learning techniques, including, but not limited to, the following classes: supervised training, semi-supervised training, unsupervised training, self-learning, reinforcement learning, federated learning, transfer learning, feature learning (including principal component and cluster analysis), multi-linear subspace learning, manifold learning, representation learning (including preliminary dictionary learning), rule-based machine learning, anomaly detection, and variations or combinations thereof. After the machine-learned model has been traced, the machine-learned model may be used by the vehicle (e.g., transmitted to the vehicle via network 1190) and / or the machine-learned model may be used by server 1178 to remotely monitor the vehicle.

[0201] In some instances, the server 1178 can receive data from the vehicles and apply the data to state-of-the-art real-time neural networks for real-time intelligent inference. The server 1178 can include deep learning supercomputers and / or dedicated AI computers powered by GPUs 1184, such as the DGX and DGX Station machines developed by NVIDIA. However, in some instances, the server 1178 can include a deep learning infrastructure that uses only CPU-powered data centers.

[0202] The deep learning infrastructure of the server 1178 may be capable of rapid real-time inference and may use that capability to evaluate and validate the health of the processor, software, and / or associated hardware within the vehicle 1100. For example, the deep learning infrastructure may receive periodic updates from the vehicle 1100, such as a sequence of images and / or objects that the vehicle 1100 has located within the sequence of images (e.g., via computer vision and / or other machine learning object classification techniques). The deep learning infrastructure may run its own neural networks to identify objects and compare them to objects identified by the vehicle 1100, and if the results do not match and the infrastructure concludes that the AI ​​within the vehicle 1100 is not functioning properly, the server 1178 may send a signal to the vehicle 1100 to infer control, notify the passenger, and command the failsafe computer of the vehicle 1100 to complete a safe parking maneuver.

[0203] For inference, the server 1178 may include a GPU 1184 and one or more programmable inference accelerators (e.g., NVIDIA's TensorRT). The combination of a GPU-powered server and inference acceleration can enable real-time responsiveness. In other instances, such as when less performance is required, servers powered by CPUs, FPGAs, and other processors may be used for inference.

[0204] The present disclosure may be described in the general context of computer code or machine usable instructions, including computer executable instructions, such as program modules, being executed by a computer or other machine, such as a personal digital assistant or other handheld device. Generally, program modules, including routines, programs, objects, components, data structures, etc., refer to code that performs particular tasks or implements particular abstract data types. The present disclosure may be implemented in a variety of configurations, including handheld devices, consumer electronics, general purpose computers, more specialized computing devices, etc. The present disclosure may also be implemented in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network.

[0205] As used herein, the statement "and / or" with respect to two or more elements should be interpreted to mean only one element or a combination of elements. For example, "element A, element B, and / or element C" may include element A only, element B only, element C only, elements A and element B, elements A and element C, elements B and element C, or elements A, B, and C. In addition, "at least one of element A or element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. Furthermore, "at least one of element A and element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B.

[0206] The subject matter of the present disclosure has been described with specificity to meet statutory requirements. However, the description itself is not intended to limit the scope of the present disclosure. Instead, the inventors intend that the claimed subject matter may be implemented in other ways, including different steps or combinations of steps similar to those described in this document, in conjunction with other current or future technologies. Furthermore, although the terms "step" and / or "block" may be used herein to connote different elements of the method used, these terms should not be construed as implying any particular order between the various steps disclosed herein unless and when the order of the individual steps is explicitly described.

Claims

1. identifying a region in a first image corresponding to an object having a first background in the first image; determining image data representative of the object based at least on the region of the object; generating a second image including the object with the second background based at least on integrating the object with a second background using the image data; training at least one neural network to perform a predictive task using the second image; selecting a view of the object in an environment; generating the first image based at least on rasterizing a three-dimensional capture of the object in the environment from the view; A method comprising:

2. 2. The method of claim 1 , wherein the identifying step comprises determining at least a first segment of the first image that corresponds to the object and at least a second segment of the first image that corresponds to the first background based at least on performing image segmentation on the first image.

3. The method of claim 1 , wherein the training of the at least one neural network comprises classifying one or more poses of the object.

4. The method of claim 1 , wherein the determining the image data comprises: generating a mask based at least on the region in the first image; and applying the mask to the first image.

5. 2. The method of claim 1, wherein the determining step of the image data includes generating a mask based at least on the region in the first image, and the generating step of the second image is based at least on performing one or more of dilation or erosion on a portion of the mask that corresponds to the object.

6. 2. The method of claim 1, wherein the determining step of the image data includes generating a mask based at least on the region in the image, and the generating step is based at least on blurring at least a portion of a boundary of the mask that corresponds to the object.

7. The method of claim 1 , wherein said generating said second image comprises modifying a hue of said object.

8. 10. The method of claim 1, further comprising generating a third image including the object with a third background based at least on integrating the object with a third background, and wherein the training of the at least one neural network further uses the third image.

9. The method of claim 1 , wherein the merging comprises seamlessly blending the object with the second background.

Images