On-device artificial intelligence video search

Abstract

A computer-implemented method for on-device video query and search using an artificial neural network (ANN) includes receiving, by the ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The ANN generates, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. A correlation between the search query and the subtitle information is determined at the mobile device by the ANN based on the first and second representations. The ANN predicts, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to Indian Patent Application No. 202241011422, entitled "ON-DEVICE ARTIFICIAL INTELLIGENCE VIDEO SEARCH", filed on March 3, 2022, the entire disclosure of which is expressly incorporated herein by reference.

[0002] Aspects of the present disclosure relate generally to neural networks, and more specifically, to on-device video search using artificial neural networks. [Background technology]

[0003]

[0003] An artificial neural network may comprise an interconnected group of artificial neurons (e.g., neuron models). An artificial neural network may be a computational device or may be represented as a method to be executed by a computational device. A convolutional neural network is a type of feed-forward artificial neural network. A convolutional neural network may include a collection of neurons, each with a receptive field, that collectively tile the input space. Convolutional neural networks (CNNs), such as deep convolutional neural networks (DCNs), have numerous applications. In particular, these neural network architectures are used in a variety of technologies, such as image recognition, pattern recognition, speech recognition, autonomous driving, and other classification tasks.

[0004]

[0004] Edge devices such as smartphones or other mobile devices are widely used to consume media, such as music or videos. Given the many useful applications of neural networks, there is an increasing demand for their use on edge devices. Searching for specific content within a video, song, or other sequence is a common task for users. For example, often a user may want to play, for example, a favorite or memorable scene from a movie, an important event (e.g., a goal), a dialogue, or a conversation in a video, without watching the entire movie or video. However, automatically searching for such events is cumbersome, time-consuming, and computationally expensive in terms of power. This is especially acute in resource-limited devices such as mobile devices. Summary of the Invention

[0005]

[0005] The present disclosure is set out in the independent claims, some aspects of which are set out in the dependent claims.

[0006] In one aspect of the disclosure, a computer-implemented method for searching videos on a mobile device using an artificial neural network (ANN) includes receiving, by the ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The computer-implemented method also includes generating, by the ANN, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. The computer-implemented method additionally includes determining, by the ANN, at the mobile device, a correlation based on the first and second representations. The computer-implemented method further includes predicting, by the ANN, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

[0007]

[0007] Another aspect of the present disclosure is directed to an apparatus including means for receiving, by an ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The apparatus also includes means for generating, by the ANN, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. In addition, the apparatus includes means for determining, by the ANN, at the mobile device, a correlation based on the first and second representations. The apparatus further includes means for predicting, by the ANN, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

[0008]

[0008] In another aspect of the present disclosure, a non-transitory computer-readable medium is presented. The non-transitory computer-readable medium has recorded thereon program code for searching for videos on a mobile device using an artificial neural network (ANN). The program code is executed by a processor and includes program code for receiving, by the ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The program code also includes program code for generating, by the ANN, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. The program code additionally includes program code for determining, by the ANN, at the mobile device, a correlation based on the first representation and the second representation. The program code further includes program code for predicting, by the ANN, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

[0009]

[0009] Another aspect of the disclosure is directed to an apparatus for searching videos on a mobile device using an artificial neural network (ANN). The apparatus includes a memory and one or more processors coupled to the memory. The processor(s) are configured to receive, by the ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The processor(s) are also configured to generate, by the ANN, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. The processor(s) are additionally configured to determine, by the ANN, at the mobile device, a correlation based on the first and second representations. The processor(s) are further configured to predict, by the ANN, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

[0010]

[0010] Additional features and advantages of the present disclosure are described below. Those skilled in the art will appreciate that the present disclosure may be easily utilized as a basis for modifying or designing other structures for carrying out the same purpose of the present disclosure. Those skilled in the art will also appreciate that such equivalent constructions do not depart from the teachings of the present disclosure as set forth in the appended claims. The novel features believed to be characteristic of the present disclosure, both in terms of its organization and method of operation, together with further objects and advantages, will be better understood from the following description when considered in conjunction with the accompanying figures. It should be expressly understood, however, that each of the figures is provided for illustration and description only, and is not intended to define the scope of the present disclosure. [Brief description of the drawings]

[0011]

[0011] The features, nature, and advantages of the present disclosure will become more apparent from the detailed description set forth below when read in conjunction with the drawings in which like reference characters identify correspondingly throughout. [Figure 1]

[0012] FIG. 1 illustrates an example implementation of a neural network using a system on a chip (SOC) that includes a general-purpose processor, in accordance with some aspects of the present disclosure. [Figure 2A]

[0013] FIG. 1 illustrates a neural network according to an aspect of the present disclosure. [Figure 2B] FIG. 1 illustrates a neural network according to an aspect of the present disclosure. [Figure 2C] FIG. 1 illustrates a neural network according to an aspect of the present disclosure. [Figure 2D]

[0014] FIG. 1 illustrates an exemplary deep convolutional network (DCN), in accordance with aspects of the present disclosure. [Diagram 3]

[0015] FIG. 1 is a block diagram illustrating an example deep convolutional network (DCN), in accordance with aspects of the present disclosure. [Figure 4]

[0016] FIG. 1 is a block diagram illustrating an example software architecture that may modularize artificial intelligence (AI) functionality, according to aspects of the present disclosure. [Diagram 5]

[0017] 1 is a high-level block diagram illustrating an example process for on-device query and retrieval of videos according to an aspect of the disclosure. [Figure 6A]

[0018] 1 is a block diagram illustrating an example architecture for video query and search, according to an aspect of the present disclosure. [Figure 6B]

[0019] FIG. 1 is a block diagram illustrating an example on-device neural network, according to an aspect of the present disclosure. [Figure 7]

[0020] FIG. 1 is a flow diagram illustrating an example process for video query and retrieval using an artificial neural network, according to an aspect of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012]

[0021] The detailed description of the invention described below with reference to the accompanying drawings illustrates various configurations and does not represent the only configurations in which the described concepts may be practiced. The detailed description of the invention includes specific details intended to provide a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well-known structures and components are shown in block diagram form to avoid obscuring such concepts.

[0013]

[0022] Based on the teachings, those skilled in the art should understand that the scope of the present disclosure is intended to encompass any aspect of the present disclosure, whether implemented independently of or in combination with any other aspect of the present disclosure. For example, an apparatus may be implemented or a method may be practiced using any number of the aspects described. In addition, the scope of the present disclosure is intended to encompass such an apparatus or method that is practiced using other structures, functions, or structures and functions in addition to or other than the various aspects of the present disclosure described. It should be understood that any aspect of the present disclosure disclosed may be embodied by one or more elements of a claim.

[0014]

[0023] The word "exemplary" is used to mean "serving as an example, instance, or illustration." Any aspect described as "exemplary" is not necessarily to be construed as preferred or advantageous over other aspects.

[0015]

[0024] Although specific aspects are described, many variations and permutations of these aspects fall within the scope of the present disclosure. Although some benefits and advantages of the preferred aspects are described, the scope of the present disclosure is not limited to any particular benefit, application, or purpose. Rather, the aspects of the present disclosure are intended to be broadly applicable to different technologies, system configurations, networks, and protocols, some of which are illustrated by way of example in the figures and the following description of the preferred aspects. The detailed description and drawings are merely illustrative of the present disclosure, rather than limiting, the scope of the present disclosure being defined by the appended claims and their equivalents.

[0016]

[0025] As described, searching for specific content in a video is a common task for users. Often, a user may want to play, for example, a favorite or memorable scene from a movie, an important event (e.g., a goal), a dialogue, or a conversation in a video, without watching the entire movie or video. However, automatically searching for such events is cumbersome, time-consuming, and computationally expensive in terms of power. This is especially acute in resource-limited devices such as mobile devices.

[0017]

[0026] Conventional techniques for video search include enhancing semantic understanding of video frames, which involves searching each of the video frames to find the correct video moment to be shown in the search. To generate faster results, some conventional techniques also include pre-processing all video frames and storing the learning in a cached corpus / database. However, semantic understanding of video frames is a challenging task. In addition, such conventional techniques are time-consuming and a significant amount of time is spent searching a video. Furthermore, conventional techniques are computationally expensive, which is significant in resource-limited devices such as smartphones or other mobile devices.

[0018]

[0027] To address these and other challenges, aspects of the present disclosure are directed to on-device query and search for videos. An artificial neural network can process a search query from a user, subtitle information from the video, and generate predictions indicating portions of the video that are most likely to contain the searched content. In some aspects, the artificial neural network can generate a list of N predictions of likely events that match the search query or are likely to contain the searched content (e.g., if the search query describes an event that occurs multiple times), where N is an integer.

[0019]

[0028] In some aspects, the prediction may include a start time and an end time to identify a portion that includes the searched content. The predicted portion may be displayed as text or a timestamp (e.g., a timeline position) in a video or media player, allowing a user to easily navigate to the predicted portion to play the searched content, or in some aspects, the mobile device may automatically play the predicted portion. The processing and prediction may occur on-device. On-device may refer to processing and prediction without the assistance of a cloud or remote computing.

[0020]

[0029] 1 illustrates an exemplary implementation of a system-on-a-chip (SOC) 100, which may include a central processing unit (CPU) 102 or a multi-core CPU configured to search a video using an artificial neural network (e.g., a neural end-to-end network). Variables (e.g., neural signals and synaptic weights), system parameters associated with a computational device (e.g., a neural network with weights), delays, frequency bin information, and task information may be stored in a memory block associated with a neural processing unit (NPU) 108, in a memory block associated with the CPU 102, in a memory block associated with a graphics processing unit (GPU) 104, in a memory block associated with a digital signal processor (DSP) 106, in a memory block 118, or distributed across multiple blocks. Instructions executed in the CPU 102 may be loaded from a program memory associated with the CPU 102 or from the memory block 118.

[0021]

[0030] The SOC 100 may also include a GPU 104, a DSP 106, a connectivity block 110 that may include fifth generation (5G) connectivity, fourth generation long term evolution (4G LTE) connectivity, Wi-Fi connectivity, USB connectivity, Bluetooth connectivity, etc., as well as additional processing blocks adapted to specific functions, such as a multimedia processor 112 that may detect and recognize gestures. In one implementation, the NPU 108 is implemented in the CPU 102, the DSP 106, and / or the GPU 104. The SOC 100 may also include a sensor processor 114, image signal processors (ISPs) 116, and / or a navigation module 120 that may include a global positioning system.

[0022]

[0031] The SOC 100 may be based on an ARM instruction set. In an aspect of the disclosure, the instructions loaded into the general purpose processor 102 may include code for receiving, by the ANN, a video and a search query. The video includes a sequence of frames and associated subtitle information. The general purpose processor 102 may also include code for generating, by the ANN, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information. The general purpose processor 102 may additionally include code for determining, by the ANN, at the mobile device, a correlation based on the first and second representations. The general purpose processor 102 may further include code for predicting, by the ANN, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation.

[0023]

[0032] Deep learning architectures may perform object recognition tasks by learning to represent inputs at successively higher levels of abstraction within each layer, thereby building useful feature representations of the input data. In this way, deep learning addresses a major bottleneck of traditional machine learning. Prior to the advent of deep learning, machine learning approaches to object recognition problems may have relied heavily on human-designed features, possibly in combination with shallow classifiers. A shallow classifier may be, for example, a two-class linear classifier that can compare a weighted sum of feature vector components to a threshold to predict which class an input belongs to. Human-designed features may be templates or kernels that are adapted to a particular problem domain by an engineer with domain expertise. In contrast, deep learning architectures may learn, but through training, to represent features that are similar to those that a human engineer could design. Furthermore, deep networks may learn to represent and recognize new types of features that humans may not have thought of.

[0024]

[0033] Deep learning architectures may learn a hierarchy of features. When presented with visual data, for example, a first layer may learn to recognize relatively simple features such as edges in the input stream. In another example, when presented with auditory data, the first layer may learn to recognize spectral power at specific frequencies. A second layer, taking as input the output of the first layer, may learn to recognize combinations of features such as simple shapes in the case of visual data, or combinations of sounds in the case of auditory data. For example, higher layers may learn to represent complex shapes in visual data or words in auditory data. Even higher layers may learn to recognize common visual objects or spoken phrases.

[0025]

[0034] Deep learning architectures may work particularly well when applied to problems that have a natural hierarchical structure. For example, classification of electric vehicles may benefit from first learning to recognize wheels, windshields, and other features. These features may be combined in different ways at higher layers to recognize cars, trucks, and planes.

[0026]

[0035] Neural networks may be designed with various connectivity patterns. In feedforward networks, each neuron in a given layer communicates with neurons in a higher layer, so that information is passed from lower layers to higher layers. As described above, hierarchical representations may be constructed in successive layers of a feedforward network. Neural networks may also have recurrent or feedback (also called top-down) connections. In recurrent connections, the output from a neuron in a given layer may be transmitted to another neuron in the same layer. Recurrent architectures may be useful in recognizing patterns across two or more of the input data chunks delivered in sequence to the neural network. Connections from neurons in a given layer to neurons in a lower layer are called feedback (or top-down) connections. Networks with many feedback connections may be useful when recognition of high-level concepts can help discriminate certain low-level features of the input.

[0027]

[0036] The connections between layers of a neural network may be fully connected or locally connected. FIG. 2A shows an example of a fully connected neural network 202. In the fully connected neural network 202, a neuron in a first layer may transmit its output to every neuron in a second layer, so that each neuron in the second layer receives input from every neuron in the first layer. FIG. 2B shows an example of a locally connected neural network 204. In the locally connected neural network 204, a neuron in a first layer may be connected to a limited number of neurons in the second layer. More generally, the locally connected layers of the locally connected neural network 204 may be configured such that each neuron in a layer has the same or similar connectivity pattern, but with connection strengths that may have different values ​​(e.g., 210, 212, 214, and 216). Because higher layer neurons in a given region may receive inputs that are tuned through training to the properties of a constrained subset of all inputs to the network, the connectivity patterns of local connections may give rise to spatially distinct receptive fields within the higher layers.

[0028]

[0037] An example of a locally connected neural network is a convolutional neural network. Figure 2C shows an example of a convolutional neural network 206. The convolutional neural network 206 may be configured such that the connection strengths associated with the inputs for each neuron in the second layer are shared (e.g., 208). Convolutional neural networks may be suitable for problems where the spatial location of the inputs is meaningful.

[0029]

[0038] One type of convolutional neural network is the deep convolutional network (DCN). Figure 2D shows a detailed example of a DCN 200 designed to recognize visual features from images 226 input from an image capture device 230, such as an on-board camera. The DCN 200 in this example may be trained to identify traffic signs and numbers on traffic signs. Of course, the DCN 200 may be trained for other tasks, such as identifying lane markings or identifying traffic signals.

[0030]

[0039] The DCN 200 may be trained using supervised learning. During training, the DCN 200 may be presented with an image, such as an image 226 of a speed limit sign, and then a forward pass may be computed to generate the output 222. The DCN 200 may include a feature extraction section and a classification section. Upon receiving the image 226, the convolution layer 232 may apply a convolution kernel (not shown) to the image 226 to generate the first set of feature maps 218. As an example, the convolution kernel for the convolution layer 232 may be a 5×5 kernel that generates a 28×28 feature map. In this example, four different convolution kernels were applied to the image 226 in the convolution layer 232 because four different feature maps are generated in the first set of feature maps 218. A convolution kernel may also be referred to as a filter or a convolution filter.

[0031]

[0040] The first set of feature maps 218 may be subsampled by a max pooling layer (not shown) to generate a second set of feature maps 220. The max pooling layer reduces the size of the first set of feature maps 218. That is, the size of the second set of feature maps 220, such as 14×14, is smaller than the size of the first set of feature maps 218, such as 28×28. The reduced size provides similar information to subsequent layers while reducing memory consumption. The second set of feature maps 220 may be further convolved through one or more subsequent convolutional layers (not shown) to generate one or more subsequent sets of feature maps (not shown).

[0032]

[0041] In the example of Figure 2D, the second set of feature maps 220 are convolved to generate a first feature vector 224. Furthermore, the first feature vector 224 is further convolved to generate a second feature vector 228. Each feature in the second feature vector 228 may include a number corresponding to a possible feature of the image 226, such as "sign", "60", and "100". A softmax function (not shown) may convert the numbers in the second feature vector 228 into probabilities. Thus, the output 222 of the DCN 200 is the probability that the image 226 contains one or more features.

[0033]

[0042] In this example, the probability in output 222 for "sign" and "60" is higher than the probability for other of the outputs 222 such as "30", "40", "50", "70", "80", "90", and "100". Before training, the output 222 generated by DCN 200 may be inaccurate. Thus, an error may be calculated between the output 222 and a target output. The target output is the ground truth of image 226 (e.g., "sign" and "60"). The weights of DCN 200 may then be adjusted so that the output 222 of DCN 200 is more closely aligned with the target output.

[0034]

[0043] To adjust the weights, the learning algorithm may calculate a gradient vector for the weights. The gradient can indicate the amount by which the error would increase or decrease if the weights were adjusted. In the top layer, the gradient can correspond directly to the values ​​of the weights connecting the activated neurons in the penultimate layer to the neurons in the output layer. In lower layers, the gradient can depend on the values ​​of the weights and the calculated error gradients of the upper layers. The weights can then be adjusted to reduce the error. This method of adjusting weights is sometimes called "backpropagation" because it involves a "backward pass" through the neural network.

[0035]

[0044] In practice, the error gradient of the weights may be calculated over a small number of examples such that the calculated gradient approximates the true error gradient. This approximation may be referred to as stochastic gradient descent. Stochastic gradient descent may be repeated until the achievable error rate of the entire system stops decreasing or until the error rate reaches a target level. After training, the DCN may be presented with a new image and a forward pass through the network may give an output 222, which may be considered an estimate or prediction of the DCN.

[0036]

[0045] Deep belief networks (DBNs) are probabilistic models with multiple layers of hidden nodes. DBNs may be used to extract a hierarchical representation of a training dataset. DBNs may be obtained by stacking layers of Restricted Boltzmann Machines (RBMs). RBMs are a type of artificial neural network that can learn a probability distribution over a set of inputs. RBMs are frequently used in unsupervised learning because they can learn a probability distribution without information about the class into which each input should be categorized. Using a hybrid unsupervised-supervised paradigm, the lower RBM of the DBN can be trained in an unsupervised manner and can function as a feature extractor, and the upper RBM can be trained in a supervised manner (on the joint distribution of inputs from previous layers and the target class) and can function as a classifier.

[0037]

[0046] Deep convolutional networks (DCNs) are networks of convolutional networks constructed with additional pooling and normalization layers. DCNs have achieved state-of-the-art performance for many tasks. DCNs can be trained using supervised learning, where both the input and output targets are known for a large number of examples and are used to modify the network weights by using gradient descent.

[0038]

[0047] The DCN may be a feedforward network. In addition, as described above, connections from neurons in a first layer of the DCN to groups of neurons in the next higher layer are shared across neurons in the first layer. The feedforward and shared connections of the DCN can be exploited for high speed processing. The computational burden of the DCN may be much smaller than that of a similarly sized neural network that includes, for example, recurrent or feedback connections.

[0039]

[0048] The processing of each layer of the convolutional network may be considered as a spatially invariant template or basis projection. If the input is initially decomposed into multiple channels, such as the red, green, and blue channels of a color image, the convolutional network trained on that input may be considered as three-dimensional, with two spatial dimensions along the axes of the image and a third dimension capturing color information. The output of the convolutional connections may be considered to form a feature map in the subsequent layer, with each element of the feature map (e.g., 220) receiving input from a range of neurons in the previous layer (e.g., feature map 218) and from each of multiple channels. The values ​​in the feature map may be further processed using nonlinearities such as rectification, max(0,x), etc. Values ​​from neighboring neurons may be further pooled, corresponding to downsampling, to provide additional local invariance and dimensionality reduction. Normalization, corresponding to whitening, may also be applied through lateral inhibition between neurons in the feature map.

[0040]

[0049] The performance of deep learning architectures can improve as more labeled data points become available or as computational power increases. Modern deep neural networks are routinely trained with computational resources thousands of times greater than those available to a typical researcher only 15 years ago. New architectures and training paradigms can further improve deep learning performance. Rectified linear units may reduce the training problem known as vanishing gradients. New training techniques may reduce overfitting and therefore allow larger models to achieve better generalization. Encapsulation techniques may extract data within a given receptive field, further improving overall performance.

[0041]

[0050] 3 is a block diagram illustrating a deep convolutional network 350. The deep convolutional network 350 may include multiple different types of layers based on connectivity and weight sharing. As shown in FIG. 3, the deep convolutional network 350 includes convolutional blocks 354A, 354B. Each of the convolutional blocks 354A, 354B may be configured with a convolutional layer (CONV) 356, a normalization layer (LNorm) 358, and a max pooling layer (MAX POOL) 360.

[0042]

[0051] The convolution layer 356 may include one or more convolution filters that may be applied to the input data to generate feature maps. Although only two of the convolution blocks 354A, 354B are shown, the disclosure is not so limited and instead any number of convolution blocks 354A, 354B may be included in the deep convolution network 350 according to design preferences. The normalization layer 358 may normalize the output of the convolution filters. For example, the normalization layer 358 may provide whitening or lateral inhibition. The max pooling layer 360 may provide downsampling aggregation across the space for local invariance and dimensionality reduction.

[0043]

[0052] For example, a parallel filter bank of a deep convolutional network may be loaded onto the CPU 102 or GPU 104 of the SOC 100 to achieve high performance and low power consumption. In alternative embodiments, the parallel filter bank may be loaded onto the DSP 106 or ISP 116 of the SOC 100. In addition, the deep convolutional network 350 may access other processing blocks that may be present on the SOC 100, such as the sensor processor 114 and navigation module 120, which are dedicated to sensors and navigation, respectively.

[0044]

[0053] The deep convolutional network 350 may also include one or more fully connected layers 362 (FC1 and FC2). The deep convolutional network 350 may further include a logistic regression (LR) layer 364. Between each layer 356, 358, 360, 362, 364 of the deep convolutional network 350, there are weights (not shown) that are to be updated. The output of each of the layers (e.g., 356, 358, 360, 362, 364) may serve as an input of a subsequent one of the layers (e.g., 356, 358, 360, 362, 364) in the deep convolutional network 350 to learn a hierarchical feature representation from the input data 352 (e.g., image, audio, video, sensor data, and / or other input data) provided initially to the convolution block 354A. The output of the deep convolutional network 350 is a classification score 366 for the input data 352. The classification score 366 may be a set of probabilities, where each probability is the probability that the input data contains a feature from the set of features.

[0045]

[0054] 4 is a block diagram illustrating an example software architecture 400 that may modularize artificial intelligence (AI) functions. Using this architecture, applications may be designed that can cause various processing blocks (e.g., CPU 422, DSP 424, GPU 426, and / or NPU 428) of SOC 420 to support adaptive rounding as disclosed for post-training quantization for AI applications 402 according to aspects of the present disclosure.

[0046]

[0055] The AI ​​application 402 may be configured to invoke functions defined in user space 404, which may, for example, perform scene detection and recognition indicative of the location in which the device is currently operating. The AI ​​application 402 may configure microphones and cameras differently depending on, for example, whether the scene to be recognized is an office, an auditorium, a restaurant, or an outdoor setting such as a lake. The AI ​​application 402 may make requests to compiled program code associated with libraries defined in the AI ​​function application programming interface (API) 406. The requests may ultimately rely on the output of a deep neural network configured to provide an inference response based on, for example, the video and positioning data.

[0047]

[0056] A runtime engine 408, which may be compiled code of a runtime framework, may be further accessible to the AI ​​application 402. The AI ​​application 402 may cause the runtime engine to request inference, for example, at specific time intervals or triggered by an event detected by the application's user interface. The runtime engine may then signal an operating system in the operating system (OS) space 410, such as a kernel 412 running on the SOC 420, when it is required to provide an inference response. The operating system may then cause successive relaxations of quantization to be performed on the CPU 422, the DSP 424, the GPU 426, the NPU 428, or some combination thereof. The CPU 422 may be accessed directly by the operating system, and the other processing blocks may be accessed through drivers, such as drivers 414, 416, or 418 for the DSP 424, the GPU 426, or the NPU 428, respectively. In an illustrative example, the deep neural network may be configured to run on a combination of processing blocks such as CPU 422, DSP 424, and GPU 426, or may run on NPU 428.

[0048]

[0057] An application 402 (e.g., an AI application) may be configured to invoke functions defined in user space 404, which may, for example, perform scene detection and recognition indicative of the location in which the device is currently operating. The application 402 may configure the microphone and camera differently depending on, for example, whether the scene to be recognized is an office, an auditorium, a restaurant, or an outdoor setting such as a lake. The application 402 may make a request to compiled program code associated with a library defined in SceneDetect application programming interface (API) 406 to provide an estimate of the current scene. The request may ultimately rely on the output of a differential neural network configured to provide a scene estimate based on, for example, video and positioning data.

[0049]

[0058] A runtime engine 408, which may be compiled code of a runtime framework, may also be accessible to the application 402. The application 402 may have the runtime engine request a scene estimate, for example, at a particular time interval or triggered by an event detected by the application's user interface. When the scene has been estimated, the runtime engine may send a signal to an operating system 410, such as a kernel 412, running on the SOC 420. The operating system 410 may then cause the calculations to be performed on a CPU 422, a DSP 424, a GPU 426, an NPU 428, or some combination thereof. The CPU 422 may be accessed directly by the operating system, and other processing blocks may be accessed through drivers, such as drivers 414-418 for the DSP 424, the GPU 426, or the NPU 428. In an illustrative example, the differential neural network may be configured to run on a combination of processing blocks, such as the CPU 422 and the GPU 426, or may run on the NPU 428.

[0050]

[0059] As described, aspects of the disclosure are directed to on-device query and search for videos. An artificial neural network can process a search query from a user, subtitle information from the video, and generate predictions indicating portions of the video that are most likely to contain the searched content. In some aspects, the artificial neural network can generate a list of N predictions of likely events that match the search query or are likely to contain the searched content (e.g., if the search query describes an event that occurs multiple times), where N is an integer.

[0051]

[0060] In some aspects, the prediction may include a start time and an end time to identify the portion that includes the searched content. The predicted portion may be displayed as text or a timestamp (e.g., a timeline position) in a video or media player. As a result, a user can more easily navigate to the predicted portion to play the searched content, or in some aspects, the mobile device can automatically play the predicted portion. The processing and prediction may be performed on-device. On-device may refer to processing and prediction that uses the resources of the mobile device without the assistance of cloud or remote computing.

[0052]

[0061] 5 is a high-level block diagram illustrating an example process 500 for on-device query and search of videos according to an aspect of the disclosure. With reference to FIG. 5, at block 502, process 500 can receive a search query and a video via a mobile device. The search query can include, for example, a description of a scene or event in the video.

[0053]

[0062] At block 504, the search query and subtitle information associated with the video may be processed. An artificial neural network (e.g., 350 of FIG. 3) may process the search query and the subtitle information. The subtitle information may include descriptions of scenes or actions in the video. The subtitle information may be provided in a paragraph format on a frame-by-frame basis. The ANN may determine features of the search query and the subtitle information. In some aspects, the ANN may determine salient portions of the search query to search for subtitle information associated with the video to match the features.

[0054]

[0063] At block 506, the ANN can generate a prediction of a portion of the video that contains the subject material in response to the search query. The prediction can include a set of frames that are most likely to contain the scene or event described in the search query. In some aspects, the prediction can include a start time and an end time for the portion of the video that is most likely to contain the scene or event described in the search query.

[0055]

[0064] At block 508, the predicted portion of the video may be displayed on a display of the mobile device (eg, via the multimedia processor 112 of FIG. 1).

[0056]

[0065] FIG. 6A is a block diagram illustrating an example architecture 600 for video query and search, according to an aspect of the disclosure. With reference to FIG. 6A, the architecture 600 may include an input device 602. The input device 602 may include a microphone of a mobile device (e.g., the sensor 114 of FIG. 1), a multimedia device (e.g., the multimedia processor 112 of FIG. 1), or an input / output device for receiving text input. In some aspects, the input device 602 may be configured to convert a speech signal or other sensor input into text input. The input device 602 may receive a search query from a user. The search query may be a description of a scene or event contained in a movie. The search query may be in the form of a word, a phrase, or a sentence.

[0057]

[0066] The architecture 600 can also receive a video input 604. The video input 604 may be, for example, a movie. The video input 604 may be received from a storage or streaming media source. The video input 604 may include a temporal sequence of frames. In some aspects, the video input 604 may include associated closed captioning (CC) information or subtitle information. Closed captioning information, sometimes referred to as subtitle information, may include a timed transcript of dialogue in a movie between characters displayed in a frame of video. In some aspects, a subtitle generator 612 is optionally included on-device and may be used to generate subtitle information for frames of the video input 604, for example, when such information is not included with the video input 604. The subtitle information may be in the form of sentences or paragraphs, for example.

[0058]

[0067] The search query and the video subtitle information can be fed to an on-device neural network 606. The on-device neural network 606 can be, for example, a Transformer Neural Network. The Transformer Neural Network is a deep learning model that uses self-attention to provide contextual information about any position in the input sequence. In some aspects, the on-device neural network 606 can be an efficiently learning encoder that classifies token replacements accurately (ELECTRA) small model. Of course, this is only one example, and other architectures such as bi-directional encoder representations from transformers (BERT), robustly optimized BERT approach (RoBERTa), XLNet, Transformer-XL, and generative pre-trained transformer (GPT) family of transformers can also be employed. The ELECTRA small model is a question-answering natural language processor (NLP). The ELECTRA small model can be pre-trained to predict answers given a query and an input paragraph.

[0059]

[0068] The on-device neural network 606 can generate a representation of the words in the search query and a representation of the words in the subtitle information. Each of the representations of the words can include context information based on each of the words in the search query and the subtitle information, respectively. The on-device neural network 606 can determine a correlation between the search query and the subtitle information based on the generated representations of the words. The on-device neural network 606 can then generate a prediction 608 of a portion of the video input 604 that includes content in response to the search query based on the correlation. The prediction 608 can indicate a portion of the video that is most likely to include the searched content. In some aspects, the prediction 608 can indicate one or more frames that include such content. For example, as shown in the example of FIG. 6A, the on-device neural network 606 can generate three candidate portions, each of which includes subtitle information that can satisfy the search query. In some aspects, the prediction 608 can also include a start time and an end time for a portion or segment of the video input 604 that includes content in response to the search query.

[0060]

[0069] The playback device 610 can display a list of the predicted portion(s) of the video input 604 indicated by a start time and an end time. Additionally, in some aspects, the playback device can navigate (e.g., fast forward) to the identified start time and begin playing the predicted portion of the video input 604 until the identified end time.

[0061]

[0070] 6B is a block diagram illustrating an example on-device neural network 606 according to an embodiment of the present disclosure. With reference to FIG. 6B, the on-device neural network 606 can include a search generator network 620 and a discriminator network 630. The search generator network 620 and the discriminator network 630 can each be configured, for example, as a transformer network. The search generator network 620 can receive a search query as an input and convert the search query into a context vector representation h SQ The context vector representation h SQ may focus on important words (e.g., nouns, verbs) in the search query. SQ may be provided to a discriminator network 630.

[0062]

[0071] The discriminator network 630 may receive subtitle information as a first input. The discriminator network 630 may divide the subtitle information associated with each frame of the video input 604 into a context vector representation h SI In some aspects, the discriminator network 630 may map the context vector representation h SI can be generated.

[0063]

[0072] The discriminator network 630 discriminates the context vector representation h SQ The discriminator network 630 may also receive the context vector representation h SQ Let h be the context vector representation. SI Compared with the context vector representation h SQ and the context vector representation h SI That is, the discriminator network 630 can be trained to distinguish words in the data (e.g., only in the search query) from words contained in the subtitle information. SQ and the context vector representationSI Based on the correlation between, the discriminator network 630 can generate a prediction (e.g., 608) of whether the search query matches the subtitle information and, therefore, whether the corresponding portion of the video input 604 contains content that satisfies the search query.

[0064]

[0073] FIG. 7 is a flow diagram illustrating an example process 700 for video query and search using an artificial neural network (ANN) according to an aspect of the disclosure. As shown in FIG. 7, at block 702, the process 700 receives, by the ANN, a video including a sequence of frames and associated subtitle information and a search query. For example, as described with reference to FIG. 6A, the input device 602 can receive the search query from a user. The search query can be a description of a scene or event contained in the movie. The search query can be in the form of a word, phrase, or sentence. The architecture 600 can also receive a video input 604. The video input 604 can be, for example, a movie. The video input 604 can be received from a storage or streaming media source. The video input 604 can include a temporal sequence of frames. In some aspects, the video input 604 can include associated closed captioning (CC) information or subtitle information. Closed captioning information, sometimes referred to as subtitle information, can include a timed transcript of dialogue in the movie between characters displayed in frames of the video.

[0065]

[0074] At block 704, the process 700 generates, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information by an ANN. As described with reference to FIG. 6B, the search generator network 620 can receive the search query as an input and convert the search query into a context vector representation h SQ The context vector representation hSQ may focus on important words (e.g., nouns, verbs) in the search query. Similarly, the discriminator network 630 may receive subtitle information. The discriminator network 630 converts the subtitle information associated with each frame of the video input 604 into a context vector representation h SI can be mapped to

[0066]

[0075] At block 706, the process 700 determines, at the mobile device, by an ANN, a correlation based on the first and second representations. For example, as described with reference to FIG. 6B, the discriminator network 630 may determine a correlation based on the context vector representation h SQ Let h be the context vector representation. SI Compared with the context vector representation h SQ and the context vector representation h SI That is, the discriminator network 630 can be trained to distinguish words of the data (e.g., only in the search query) from words contained in the subtitle information. Additionally, the correlation may be based on a correspondence between the context for words of the search query and the context for words of the subtitle information.

[0067]

[0076] At block 708, the process 700 predicts, at the mobile device, a portion of the video that contains the content based on the correlation by the ANN in response to the search query. For example, as described with reference to FIG. 6B, SQ and the context vector representation h SIBased on the correlation between the search query, the discriminator network 630 can generate a prediction (e.g., 608) of whether the search query matches the subtitle information and, therefore, whether the corresponding portion of the video input 604 contains content that satisfies the search query. The prediction 608 can indicate a portion of the video that is most likely to contain the searched content. In some aspects, the prediction 608 can indicate one or more frames that contain such content. In some aspects, the prediction 608 can also include start and end times for a portion or segment of the video input 604 that contains content, responsive to the search query.

[0068]

[0077] The following numbered clauses contain implementations: 1. A computer-implemented method for searching videos on a mobile device using an artificial neural network (ANN), comprising: receiving, by the ANN, a video and a search query, the video including a sequence of frames and associated subtitle information; generating, at the mobile device, by an ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; determining, at the mobile device, by an ANN, a correlation based on the first representation and the second representation; predicting, at the mobile device, by the ANN, a portion of the video that includes the content in response to the search query based on the correlation; A computer-implemented method comprising: 2. The computer-implemented method of clause 1, wherein predicting further indicates a start time and an end time for a portion of a video that includes the content in response to the search query. 3. The computer-implemented method of clause 1 or 2, further comprising displaying the portion of the video contained within the start time until the end time. 4. The computer-implemented method of any one of clauses 1 to 3, wherein the ANN comprises a Transformer Neural Network. 5. The computer-implemented method of any one of clauses 1 to 4, further comprising generating associated subtitle information based on closed captioning information included with the video. 6. The computer-implemented method of any one of clauses 1 to 5, wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase. 7. The computer-implemented method of any one of clauses 1 to 6, wherein the search query is provided via speech input text input of a mobile device. 8. An apparatus for searching videos on a mobile device using an artificial neural network (ANN), comprising: Memory, at least one processor coupled to the memory, the at least one processor comprising: receiving, by the ANN, a video and a search query, the video including a sequence of frames and associated subtitle information; generating, at the mobile device, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information, determining, at the mobile device, by the ANN, a correlation based on the first representation and the second representation; and predicting, at the mobile device, a portion of the video that includes the content in response to the search query based on the correlation using an ANN. The apparatus is configured to: 9. The apparatus of clause 8, wherein the at least one processor is further configured to generate a prediction indicating a start time and an end time for a portion of a video that includes the content in response to the search query. 10. The apparatus of clause 8 or 9, wherein the at least one processor is further configured to display a portion of the video included in the start time until the end time. 11. The apparatus of any one of clauses 8 to 10, wherein the ANN comprises a Transformer Neural Network. 12. The apparatus of any one of clauses 8 to 11, wherein at least one processor is further configured to generate associated subtitle information based on closed captioning information included with the video. 13. The apparatus of any one of clauses 8 to 12, wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase. 14. The apparatus of any one of clauses 8 to 13, wherein the search query is supplied via speech input text input of a mobile device. 15. A non-transitory computer-readable medium having recorded thereon program code for searching videos on a mobile device using an artificial neural network (ANN), the program code being executed by a processor; program code for receiving, by an ANN, a video and a search query, the video including a sequence of frames and associated subtitle information; program code for generating, at the mobile device, with an ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; program code for determining, at the mobile device, by an ANN, a correlation based on the first representation and the second representation; program code for predicting, at the mobile device, a portion of the video that includes the content in response to the search query based on the correlation with the ANN; A non-transitory computer readable medium comprising: 16. The non-transitory computer-readable medium of claim 15, further comprising program code for generating a prediction indicating a start time and an end time for a portion of a video that includes the content in response to a search query. 17. The non-transitory computer readable medium of clause 15 or 16, wherein the program code further comprises program code for displaying a portion of the video that is included in the start time until the end time. 18. The non-transitory computer-readable medium of any one of clauses 15 to 17, wherein the ANN comprises a Transformer Neural Network. 19. The non-transitory computer-readable medium of any one of clauses 15 to 18, wherein the program code further comprises program code for generating associated subtitle information based on closed captioning information included with the video. 20. The non-transitory computer-readable medium of any one of clauses 15 to 19, wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase. 21. The non-transitory computer-readable medium of any one of clauses 15 to 20, wherein the search query is supplied via speech input text input of a mobile device. 22. An apparatus for searching videos on a mobile device using an artificial neural network (ANN), comprising: means for receiving, by the ANN, a video and a search query, the video including a sequence of frames and associated subtitle information; means for generating, in the mobile device, by an ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; means for determining, in the mobile device, by the ANN, a correlation based on the first representation and the second representation; means for predicting, in the mobile device, a portion of the video that includes the content in response to the search query based on the correlation using an ANN; An apparatus comprising: 23. The apparatus of clause 22, further comprising means for generating a prediction indicating a start time and an end time for a portion of a video that includes the content in response to the search query. 24. An apparatus as described in clause 22 or 23, further comprising means for displaying a portion of the video contained in the start time until the end time. 25. An apparatus as described in any one of clauses 22 to 24, wherein the ANN comprises a Transformer Neural Network. 26. An apparatus according to any one of clauses 22 to 25, further comprising means for generating associated subtitle information based on closed captioning information included with the video. 27. An apparatus according to any one of clauses 22 to 26, wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase. 28. The apparatus of any one of clauses 22 to 27, wherein the search query is supplied via speech input text input of a mobile device.

[0069]

[0078] In one aspect, the means for receiving, the means for generating, the means for determining, the means for predicting, and / or the means for displaying may be the CPU 102, a program memory associated with the CPU 102, the dedicated memory block 118, the fully connected layer 362, the NPU 428, and / or the routing connection processing unit 216 configured to perform the recited functions. In another configuration, the above-mentioned means may be any module or any device configured to perform the recited functions by the above-mentioned means.

[0070]

[0079] The various operations of the methods described above may be performed by any suitable means capable of performing the corresponding functions. The means may include various hardware and / or software component(s) and / or module(s), including but not limited to circuits, application specific integrated circuits (ASICs), or processors. In general, where there are operations illustrated in figures, those operations may have corresponding counterpart means-plus-function components that are similarly numbered.

[0071]

[0080] As used herein, the term "determining" encompasses a wide variety of actions. For example, "determining / determining" may include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, database, or another data structure), ascertaining, and the like. In addition, "determining" may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), and the like. Further, "determining" may include resolving, selecting, choosing, establishing, and the like.

[0072]

[0081] As used herein, a phrase referring to "at least one of" a list of items refers to any combination of those items, including single members. As an example, "at least one of a, b, or c" is intended to include a, b, c, ab, ac, bc, and abc.

[0073]

[0082] The various example logic blocks, modules and circuits described in connection with this disclosure may be implemented or performed using a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the described functions. A general purpose processor may be a microprocessor, but alternatively, the processor may be any commercially available processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0074]

[0083] The steps of a method or algorithm described in relation to the present disclosure may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in any form of storage medium known in the art. Some examples of storage media that may be used include random access memory (RAM), read only memory (ROM), flash memory, erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), registers, hard disks, removable disks, CD-ROMs, etc. A software module may comprise a single instruction or many instructions, and may be distributed across several different code segments, among different programs, across multiple storage media. A storage medium may be coupled to a processor such that the processor can read information from and write information to the storage medium. Alternatively, the storage medium may be integral to the processor.

[0075]

[0084] The methods disclosed herein include one or more steps or actions for achieving the described method. The steps and / or actions of those methods may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order of specific steps and / or actions and / or the use of those steps and / or actions may be modified without departing from the scope of the claims.

[0076]

[0085] The described functions may be implemented in hardware, software, firmware, or any combination thereof. When implemented in hardware, an exemplary hardware configuration may comprise a processing system in a device. The processing system may be implemented using a bus architecture. The bus may include any number of interconnected buses and bridges depending on the particular application of the processing system and the overall design constraints. The bus may link various circuits together, including the processor, the machine-readable medium, and the bus interface. The bus interface may be used to connect, among other things, a network adapter to the processing system via the bus. The network adapter may be used to implement signal processing functions. In some aspects, a user interface (e.g., a keypad, a display, a mouse, a joystick, etc.) may also be connected to the bus. The bus may also link various other circuits, such as timing sources, peripherals, voltage regulators, power management circuits, etc., which are well known in the art and therefore will not be described further.

[0077]

[0086] The processor may be responsible for managing the bus and general processing including executing software stored on the machine-readable medium. The processor may be implemented using one or more general-purpose processors and / or special-purpose processors. Examples include microprocessors, microcontrollers, DSP processors, and other circuitry capable of executing software. Software shall be broadly construed to mean instructions, data, or any combination thereof, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. The machine-readable medium may include, by way of example, a random access memory (RAM), a flash memory, a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), registers, a magnetic disk, an optical disk, a hard drive, or any other suitable storage medium, or any combination thereof. The machine-readable medium may be embodied in a computer program product. The computer program product may comprise packaging materials.

[0078]

[0087] In a hardware implementation, the machine-readable medium may be part of a processing system separate from the processor. However, as one skilled in the art will readily appreciate, the machine-readable medium, or any portion thereof, may be external to the processing system. By way of example, the machine-readable medium may include a transmission line, a carrier wave modulated by data, and / or a computer product separate from the device, all of which may be accessed by the processor through a bus interface. Alternatively, or in addition, the machine-readable medium, or any portion thereof, may be integrated into the processor, such as a cache and / or a general purpose register file. Although the various components discussed may be described as having a particular location, such as a local component, these components may also be configured in various ways, such as some components configured as part of a distributed computing system.

[0079]

[0088] The processing system may be configured as a general-purpose processing system having one or more microprocessors providing processor functionality and external memory providing at least a portion of the machine-readable medium, all linked together with other support circuits through an external bus architecture. Alternatively, the processing system may comprise one or more neuromorphic processors for implementing the described neuron model and model of the neural system. As another alternative, the processing system may be implemented using an application specific integrated circuit (ASIC) having a processor, bus interface, user interface, support circuits, and at least a portion of the machine-readable medium integrated into a single chip, or using one or more field programmable gate arrays (FPGAs), programmable logic devices (PLDs), controllers, state machines, gate logic, discrete hardware components, or any other suitable circuitry, or any combination of circuits capable of performing the various functions described throughout this disclosure. Those skilled in the art will recognize how to best implement the described functionality for the processing system depending on the particular application and the overall design constraints imposed on the overall system.

[0080]

[0089] The machine-readable medium may comprise several software modules. The software modules include instructions that, when executed by the processor, cause the processing system to perform various functions. The software modules may include a transmitting module and a receiving module. Each software module may reside in a single storage device or may be distributed across multiple storage devices. As an example, the software modules may be loaded into RAM from a hard drive when a triggering event occurs. During execution of the software modules, the processor may load some of the instructions into a cache to increase access speed. One or more cache lines may then be loaded into a general purpose register file for execution by the processor. When referring to the functionality of a software module below, it will be understood that such functionality is implemented by the processor upon executing instructions from that software module. It will further be understood that aspects of the present disclosure provide for improved functionality of a processor, computer, machine, or other system implementing such aspects.

[0081]

[0090] If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media, including any medium that facilitates transfer of a computer program from one place to another. A storage medium may be any available medium that can be accessed by a computer. By way of example, and not limitation, such computer-readable media may comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. In addition, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared (IR), radio, and microwave, the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blu-ray disc, where disks typically reproduce data magnetically and discs reproduce data optically using lasers. Thus, in some aspects, computer readable media may comprise non-transitory computer readable media (e.g., tangible media). Additionally, in other aspects, computer readable media may comprise transitory computer readable media (e.g., signals). Combinations of the above are also intended to be included within the scope of computer readable media.

[0082]

[0091] Accordingly, some aspects may include a computer program product for performing the operations presented herein. For example, such a computer program product may comprise a computer-readable medium having stored (and / or encoded) instructions, which are executable by one or more processors to perform the described operations. In some aspects, the computer program product may include packaging materials.

[0083]

[0092] Furthermore, it should be understood that modules and / or other suitable means for performing the described methods and techniques may be downloaded and / or otherwise obtained by a user terminal and / or base station, where applicable. For example, such devices may be coupled to a server to facilitate the transfer of means for performing the described methods. Alternatively, the various methods described may be provided via a storage means such that the user terminal and / or base station may obtain the various methods upon coupling or providing the storage means (e.g., RAM, ROM, physical storage medium such as a compact disk (CD) or floppy disk, etc.) to the device. Furthermore, any other suitable technique for providing the described methods and techniques to a device may be utilized.

[0084]

[0093] It should be understood that the claims are not limited to the precise configuration and components illustrated above. Various modifications, changes and variations may be made in the arrangement, operation and details of the methods and apparatus described above without departing from the scope of the claims.

Claims

1. 1. A computer-implemented method for searching videos on a mobile device using an artificial neural network (ANN), comprising: receiving, by the ANN, the video and a search query, the video including a sequence of frames and associated subtitle information; generating, at the mobile device, with the ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; determining, at the mobile device, by the ANN, a correlation based on the first representation and the second representation; predicting, at the mobile device, by the ANN, a portion of the video that includes content in response to the search query based on the correlation; A computer-implemented method comprising:

2. The computer-implemented method of claim 1 , wherein the predicting further indicates a start time and an end time for the portion of the video that includes the content in response to the search query.

3. The computer-implemented method of claim 2 , further comprising displaying the portion of the video contained in the start time until the end time.

4. The computer-implemented method of claim 1 , wherein the ANN comprises a Transform Neural Network.

5. The computer-implemented method of claim 1 , further comprising generating the associated subtitle information based on closed captioning information included with the video.

6. The computer-implemented method of claim 1 , wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase.

7. The computer-implemented method of claim 1 , wherein the search query is provided via speech-input text input of the mobile device.

8. 1. An apparatus for searching videos on a mobile device using an artificial neural network (ANN), comprising: Memory, at least one processor coupled to the memory, the at least one processor comprising: receiving, by the ANN, the video and a search query, the video including a sequence of frames and associated subtitle information; generating, at the mobile device, with the ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; determining, at the mobile device, by the ANN, a correlation based on the first representation and the second representation; predicting, at the mobile device, by the ANN, a portion of the video that includes content in response to the search query based on the correlation; The apparatus is configured to:

9. 10. The apparatus of claim 8, wherein the at least one processor is further configured to generate a prediction in response to the search query, the prediction indicating a start time and an end time for the portion of the video that includes the content.

10. 10. The apparatus of claim 9, wherein the at least one processor is further configured to display the portion of the video contained within the start time until the end time.

11. The apparatus of claim 8 , wherein the ANN comprises a Transformer Neural Network.

12. The apparatus of claim 8 , wherein the at least one processor is further configured to generate the associated subtitle information based on closed captioning information included with the video.

13. The apparatus of claim 8 , wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase.

14. The apparatus of claim 8 , wherein the search query is provided via speech-input text input of the mobile device.

15. 1. A non-transitory computer readable medium having recorded thereon program code for searching videos on a mobile device using an artificial neural network (ANN), the program code being executed by a processor, the non-transitory computer readable medium comprising: program code for receiving, by the ANN, the video and a search query, the video including a sequence of frames and associated subtitle information; program code for generating, at the mobile device, with the ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; program code for determining, at the mobile device, by the ANN, a correlation based on the first representation and the second representation; program code for predicting, at the mobile device, by the ANN, a portion of the video that includes content in response to the search query based on the correlation; A non-transitory computer readable medium comprising:

16. 20. The non-transitory computer-readable medium of claim 15, wherein the program code further comprises program code for generating a prediction in response to the search query, the prediction indicating a start time and an end time for the portion of the video that includes the content.

17. 20. The non-transitory computer readable medium of claim 16, wherein the program code further comprises program code for displaying the portion of video contained within the start time until the end time.

18. 16. The non-transitory computer-readable medium of claim 15, wherein the ANN comprises a Transformer Neural Network.

19. 20. The non-transitory computer readable medium of claim 15, wherein the program code further comprises program code for generating the associated subtitle information based on closed captioning information included with the video.

20. The non-transitory computer-readable medium of claim 15 , wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase.

21. 20. The non-transitory computer-readable medium of claim 15, wherein the search query is provided via speech-input text input of the mobile device.

22. 1. An apparatus for searching videos on a mobile device using an artificial neural network (ANN), comprising: means for receiving, by the ANN, the video and a search query, the video including a sequence of frames and associated subtitle information; means for generating, at the mobile device, with the ANN, a first representation for a first set of words in the search query and a second representation for a second set of words in the subtitle information; means for determining, in the mobile device, by the ANN, a correlation based on the first representation and the second representation; means for predicting, at the mobile device, a portion of the video that includes content in response to the search query based on the correlation, by the ANN; An apparatus comprising:

23. 23. The apparatus of claim 22, further comprising means for generating a prediction in response to the search query, the prediction indicating a start time and an end time for the portion of the video that includes the content.

24. 24. The apparatus of claim 23, further comprising: means for displaying the portion of video contained within the start time until the end time.

25. The apparatus of claim 22 , wherein the ANN comprises a Transformer Neural Network.

26. 23. The apparatus of claim 22, further comprising means for generating the associated subtitle information based on closed captioning information included with the video.

27. The apparatus of claim 22 , wherein the search query includes one or more of a description of a scene, an event, a word, or a phrase.

28. 23. The apparatus of claim 22, wherein the search query is provided via speech-input text input of the mobile device.

Images