101 results about "Object label" patented technology

The invention provides a multi-granularity open vocabulary query method based on an object-level lossless Gaussian field, and the method comprises the steps: introducing an object-level Gaussian field with a global consistency codebook, rendering a learnable semantic tag vector in the Gaussian field back to a corresponding object tag; the direct mapping between the label and the corresponding uncompressed high-dimensional feature is established through the code book, so that the semantic feature of any dimension is supported, additional compression is not needed, and the understanding capability on an object is remarkably improved; according to the method, wide quantitative and qualitative evaluation is carried out in a plurality of scenes, excellent performance in the aspects of object level zero sample segmentation and open vocabulary understanding is shown, the highest precision is particularly achieved in object-component hierarchical retrieval, and meanwhile multi-granularity scene editing is supported.

The embodiment of the invention provides a multi-modal large model training method and device, a target detection method and device and electronic equipment, and relates to the technical field of artificial intelligence, and the multi-modal large model training method comprises the steps: obtaining a plurality of sample images and a first sample description text of an object labeling box in the image; utilizing the first large language model and the first text prompt to extract a category name and an attribute of an object described by the first sample description text corresponding to each sample image, and combining at least one of the obtained category name and the attribute to obtain a second sample description text; determining whether each object labeling box in the sample image is matched with each attribute of a plurality of corresponding second sample description texts or not; constructing a sample question containing the second sample description text and a corresponding sample answer in a thinking chain form to obtain a question and answer sample pair; and training the multi-modal large model by using the constructed question and answer sample pair. The complexity of descriptive target detection can be reduced, and the detection efficiency is improved.

The present disclosure relates to the fields of computer technologies and cloud computing technologies, and provides a resource scheduling method, an electronic device, and a computer-readable storage medium. The method comprises: acquiring instance performance data, wherein the instance performance data is performance index data which is obtained by performing performance data acquisition on a plurality of cloud server instances created on at least one physical server; generating label information on the basis of the instance performance data, wherein the label information comprises target object labels and instance labels, the target object labels are used for describing target object behavior attributes, and the instance labels are used for describing instance performance; determining a target resource inventory by means of the label information; and using the label information and the target resource inventory to perform resource scheduling to obtain a scheduling result. The present disclosure solves the technical problems in the related art of resource waste and affected user service operation caused by the use of subsequent governance solutions such as performance throttling and online splitting failing to completely avoid the occurrence of instance performance contention events.

The invention discloses an object group determination method and device, electronic equipment and a storage medium. The embodiment of the invention can be applied to artificial intelligence, and the method comprises the steps: obtaining preset bitmap information and object label information of at least one target object; determining a target bitmap element corresponding to the target object in the preset bitmap information; based on preset object tag information and object tag information of each target object, setting element states of target bitmap elements corresponding to each target object, and obtaining a plurality of set target bitmap elements corresponding to the preset object tag information; aggregating the plurality of set target bitmap elements corresponding to the preset object tag information to obtain object bitmap information; and determining an object group corresponding to the preset object tag information based on the element state of each bitmap element in the object bitmap information. In the embodiment of the invention, the whole bitmap information does not need to be traversed or other data structures do not need to be used. According to the scheme, the efficiency of confirming the object group can be improved.

An automatic visual data labeling framework for heterogeneous data types is described. An example method can include obtaining scan data and video data that each depict a scene including an object. The scan data can be generated by a scanner. The video data can be generated by a camera. The method can also include generating a virtual representation of the scene in a virtual environment based on the scan data. The virtual representation can include a virtual representation subset corresponding to the object. The virtual environment can be associated with a virtual camera. The method can also include applying label data to the virtual representation subset to create a labeled virtual representation subset corresponding to the object. The method can also include applying the labeled virtual representation subset to the object depicted in the video data based on a correlation of the scanner, the camera, and the virtual camera.

The invention relates to the technical field of image-text data processing, and particularly discloses a medium-high point image scene understanding method and system based on a scene interpretation graph, and the method comprises the following steps: inputting a text prompt word and a scene image into an open set target detection model, obtaining a target detection frame and a target label which need to be used for understanding the detection scene; inputting a target detection frame and a scene image to the segmentation model, and obtaining a mask of the target detection frame; calculating the relative position and size relation of the target through the label and the mask, and obtaining a scene interpretation graph; and analyzing the scene interpretation graph through a large language model, and outputting a text result for understanding the image scene. According to the method, the generalization ability of scene understanding is improved, the data training cost is saved, the description accuracy and details are improved, and the interpretability of an understanding result is improved.

An example apparatus disclosed includes interface circuitry, machine readable instructions, and programmable circuitry to at least one of execute or instantiate the machine readable instructions to identify a first source of object label representation and a second source of object label representation, the first source or the second source including an estimated label prototype vector associated with an input text-based object query, determine a first contextualized embedding for the first source and a second contextualized embedding for the second source, and combine the first contextualized embedding and the second contextualized embedding to generate a candidate object representation associated with the input text-based object query.

The embodiments of the present disclosure relate to a video processing method and apparatus, and a device and a storage medium. The method comprises: acquiring a requirement description text for making a video template, and determining first background information, which comprises an object label and the number of video clips of a video background; on the basis of the requirement description text, the first background information and candidate material information, generating constituent element information of the video template by using a generative model, wherein the constituent element information comprises at least one piece of music information, effect resource information and subtitle information; and generating a target video template on the basis of the constituent element information.

The invention discloses a timbre object separation method and device for audio data. The method comprises the following steps: acquiring original video and audio data; if the subtitle data is received, cutting the original video and audio data according to the timestamp of the subtitle data to obtain a plurality of audio coarse slice segments; respectively extracting at least one audio fine slice segment from each audio coarse slice segment by adopting a sliding window with a preset size; extracting audio features from each audio fine slice segment and clustering the audio features, and determining an object tag corresponding to each audio fine slice segment; according to the invention, each object label is backfilled to the audio coarse slice segment, and the tone object corresponding to each audio coarse slice segment is determined, so that different tones are effectively separated in a plurality of main speaker scenes through a mode of determining object label backfilling in audio coarse cutting and fine cutting processes, the requirements in complex scenes are met, and the audio separation accuracy is improved.

The present invention provides a method and system for extracting salient regions of community images. The method includes preparing a training image set, and the testing process includes the following steps: calculating appearance-based saliency for the images in the training image set; calculating semantic features of corresponding object labels for each image in the training image set; solving the fusion coefficients of salient features; extracting the salient features of the test image and calculating the saliency map. The method and system for extracting salient regions of community images proposed by the present invention calculate, in the training stage, the salient features based on the image appearance, fuse them with the calculated semantic features of object labels, and solve the fusion parameters. In the testing stage, the fusion parameters obtained through training are used to fuse the salient features of the image appearance and the label semantic features to obtain the final saliency map.

The application relates to an object representation model training method, an object representation method, an object representation device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: inputting a first training image set into an initial object representation model to obtain object association features corresponding to training images, object features and object fusion features; obtaining an object association classification loss based on the object association features corresponding to the same training image and an object association label, obtaining an object classification loss based on the object features corresponding to the same training image and an object label, and obtaining a fusion loss based on each object fusion feature corresponding to the same training image group; adjusting model parameters of the initial object representation model based on various losses to obtain an intermediate object representation model; performing model iterative training on the intermediate object representation model as the initial object representation model until a first convergence condition is met, and obtaining a target object representation model. The method can improve the representation accuracy of features for objects.

Methods, systems, and non-transitory computer readable storage media are disclosed for generating semantic scene graphs for digital images using an external knowledgebase for feature refinement. For example, the disclosed system can determine object proposals and subgraph proposals for a digital image to indicate candidate relationships between objects in the digital image. The disclosed system can then extract relationships from an external knowledgebase for refining features of the object proposals and the subgraph proposals. Additionally, the disclosed system can generate a semantic scene graph for the digital image based on the refined features of the object / subgraph proposals. Furthermore, the disclosed system can update / train a semantic scene graph generation network based on the generated semantic scene graph. The disclosed system can also reconstruct the image using object labels based on the refined features to further update / train the semantic scene graph generation network.

Provided are a method and an apparatus for auto segmentation using a bounding box. A method for auto segmentation using a bounding box according to one embodiment of the present disclosure comprises receiving a first object image including an object labeled with a bounding box, which is a pre-learning target, learning a segmentation model by classifying an object and a background from the bounding box of the received first object image, and segmenting an object from a second object image, which is an identification target, using the learned segmentation model.

The application provides an object recognition method, device and equipment and a storage medium. The method comprises: obtaining target data; identifying an object to which the target data belongs based on a target recognition model; the target recognition model uses a first training sample of a newly added object labeled with an object identifier, so that the object identifier predicted for the first training sample is consistent with the object identifier labeled for the first training sample, and the probability distribution predicted for the first training sample approximates the probability distribution predicted for the first training sample by a basic recognition model; the basic recognition model is trained by using a second training sample of a historical object labeled with an object identifier; and an initial target recognition model is a model capable of recognizing data of the historical object. The application can train the target recognition model capable of accurately identifying the object to which the data of the newly added object belongs and accurately identifying the object to which the data of the historical object belongs by using only the training data of the newly added object.

The invention provides a game interaction control method and device and electronic equipment. The method comprises the following steps: in response to entering an object selection state, selecting a target object from virtual objects based on an aiming position of a front sight identifier; displaying object tags of at least part of the target objects on a graphical user interface; in response to a trigger operation for the object tag, setting an object state of a target object corresponding to the trigger operation; and in response to the ending of the object selection state, performing collection and settlement on at least part of the target objects based on the object states of the target objects. According to the mode, a decision space is provided in a rapid collection process, wrong collection is avoided, and game operation experience is improved.

The embodiment of the invention discloses a storage packaging operation identification method and device, electronic equipment and a storage medium, and relates to the field of storage packaging, and the method comprises the steps: collecting a to-be-identified packaging frame; detecting the to-be-identified packaging frame by using a pre-trained object detection model in the storage packaging operation identification model, and determining a target object label set and a target object bounding box set in the to-be-identified packaging frame; screening out a candidate operation description set from a pre-constructed standard operation description knowledge base according to the target object label set; and performing alignment processing on the target object bounding box set and the candidate operation description set by using a comparison language image pre-training model in a storage packaging operation identification model to obtain an identification result of the packaging operation in the to-be-identified packaging frame. By adopting the technical scheme of the embodiment of the invention, the recognition accuracy of operations with subtle differences in similar environments can be improved.

The application discloses a reference video segmentation method under complex action description of a body robot, which is based on a DINO-SAM model and constructs a reference video segmentation framework under complex text description. The framework contains three modules, namely, a DINO-SAM segmentation module, an action perception aggregation module and a text-label matching module. The DINO-SAM module utilizes the powerful segmentation capability of SAM, performs preliminary segmentation on the video through text prompts, and generates object labels. The action perception aggregation module can effectively capture object action information at different time scales and enhance the global understanding of the video. The text-label matching module enhances the consistency between the segmentation result and the text description. The method effectively improves the segmentation performance, enhances the perception and aggregation capability of the object action, realizes accurate matching between the language description and the object, improves the generalization capability and the few-shot detection capability.

The embodiment of the invention discloses a region monitoring method and device, electronic equipment and a storage medium. The region monitoring method comprises the following steps: collecting a first video stream corresponding to a target object at a first preset angle through a first camera; identifying an object label corresponding to a target object of the first video stream; when identifying that the object label is not the preset object label, collecting a second video stream of the target object at a second preset angle through a second camera; and monitoring the target object based on the first video stream and the second video stream.

The embodiment of the invention provides an object annotation modification method, a program product, equipment and a storage medium. The method comprises the steps of displaying continuous multi-frame images where an annotated object is located in a first display area of a user interface, and displaying a first-frame image where the annotated object is located in a second display area of the user interface; the image content of the image comprises a labeling object and carries labeling information; in response to an annotation information modification instruction of a to-be-modified frame input by a user, modifying annotation information of a first annotation object in the to-be-modified frame into target annotation information; the first annotation object is an annotation object of to-be-modified annotation information in the to-be-modified frame; the target annotation information is annotation information of the first annotation object in a target first frame, and the target first frame is a first frame image where the first annotation object is located. According to the method, the user quickly positions the target first frame where the first labeling object is located from the second display area, correct target labeling information is obtained from the target first frame, the modification efficiency is improved, and the modification accuracy is ensured.

The present invention discloses a tag-based content processing method, computer device, and readable storage medium, relating to the field of cloud computing technology. The method comprises the following steps: obtaining a tag tree, wherein each node in the tag tree is regarded as a content; obtaining a tag dictionary table, wherein the tag dictionary table has a plurality of entries, each entry corresponding to a folder, each entry having a tag value, a tag name, and a presentation layer; obtaining an object tag table, wherein the object tag table has a plurality of entries, each entry corresponding to a content, each entry having an object ID, a tag value, and a presentation layer; when processing the content, modifying the tag dictionary table and / or the object tag table of the content; after the modification is completed, the corresponding processing is completed. The present invention achieves sharing within seconds without changing the file / folder entity, and is suitable for private cloud devices and distributed groups composed of private cloud devices.

Provided are a method and an apparatus for auto segmentation using a bounding box. A method for auto segmentation using a bounding box according to one embodiment of the present disclosure comprises receiving a first object image including an object labeled with a bounding box, which is a pre-learning target, learning a segmentation model by classifying an object and a background from the bounding box of the received first object image, and segmenting an object from a second object image, which is an identification target, using the learned segmentation model.

The application provides a data sticking method, device, equipment and program product; the method comprises the following steps: in response to a data sticking request of a target object routed by a data sticking node, obtaining an object identifier of the target object, an object label of the target object, a first sticking data list corresponding to the object label, and a bitmap of each first to-be-stuck data in the first sticking data list; determining a second sticking data list from the first sticking data list according to the bitmap and the object identifier; the second sticking data list comprises a plurality of second to-be-stuck data; determining a sticking score of each second to-be-stuck data; determining target sticking data from the plurality of second to-be-stuck data according to the sticking score; and sending the target sticking data to the data sticking node. Through the application, the accuracy of the target sticking data can be improved while reducing the cost of data sticking.

The present application provides a collision warning method, device and electronic device for a suspended fuel dispenser, which inputs the acquired target scene image into a pre-trained object annotation model, outputs the target scene image with an object annotation box annotated with the target object; detects whether there is an overlapping area between the object annotation box and the dangerous area; if there is an overlapping area and the area between the overlapping areas is greater than a preset area threshold, generates a warning message. In this way, the target scene image is acquired in real time and input into the object annotation model, and the pruned object annotation model quickly outputs the target scene image annotated with the target object. According to the area of ​​the overlapping area between the object annotation box annotated with the target object in the target scene image and the pre-set dangerous area, it is determined whether to generate a warning message, and the collision risk at the fuel dispenser can be detected in real time and quickly, which helps to improve the efficiency and timeliness of the collision warning of the suspended fuel dispenser.

The present invention discloses a method for constructing a semantic map based on laser radar and RFID. The semantic map consists of a grid map and a semantic information library; the semantic map construction method first uses SLAM to construct a grid map, and then the mobile robot obtains RFID tag information and laser radar scanning in the environment to obtain the coordinate information of the first object or the second object in the map, and finally updates the semantic information library. The present invention has the advantages of efficient and rapid acquisition of the coordinate information of the first object or the second object in the map; improves the robot's understanding of indoor semantic information, and obtains more accurate coordinate information of the first object label; does not perform precise identification on the acquisition of the coordinate information of the object label, so that it can meet the needs of navigation tasks; uses a graphic database to store semantic information, which enables other knowledge forms of the robot to be combined with map semantic information to better understand human language and maximize the use of semantic information.

This invention provides an image restoration method, apparatus, and electronic device, applicable to the field of image processing technology. The method includes: determining the location and object labels of different objects in an image to be restored; using a pre-set image restoration method for each object label, restoring the image regions corresponding to the locations of each object label in the image to be restored, thereby obtaining a restored image; wherein, the location corresponding to each object label is the location of the object to which that label belongs in the image to be restored. This solution can improve the restoration quality of image restoration.

The application relates to an attribute classification method and device, computer equipment, a storage medium and a computer program product. The attribute classification method comprises the following steps: obtaining historical data of a target object; calculating a first attribute value corresponding to the target object according to a preset interactive object label and historical information; calculating a second attribute value corresponding to the target object according to a referral identity label carried by the target object and a pre-stored label list; and determining a target classification attribute corresponding to the target object according to the first attribute value and the second attribute value. Through the arrangement, the attribute division of the target object can be more accurate, and the target classification attribute obtained finally can be more consistent with the actual attribute of the target object.

Embodiments of the present application disclose an object label generation method and device and a computer readable storage medium. After obtaining historical operation information of at least one target object in a preset time period, the embodiments of the present application filter target historical operation information corresponding to a target content of a preset content knowledge graph from the historical operation information. Then, based on the target historical operation information and the preset content knowledge graph, the historical object label of the target object is counted. Then, according to the target content and the target historical operation information, at least one predicted object label of the target object is determined from an object label set corresponding to the preset content knowledge graph. The historical object label and the predicted object label are fused to obtain the object label of the target object. The scheme can improve the accuracy of object label generation. The embodiments of the present application can be applied to various scenes such as cloud technology, artificial intelligence, intelligent transportation, and auxiliary driving.

Methods and systems for model calibration include training (210) an object detection model to generate confidence scores using calibration that is based on confidence, correlation, and matching, with accuracy of a location of bounding boxes being used with accuracy of object labels to keep the confidence scores close to an actual probability of correctness. Object detection is performed (230) on an image using the object detection model to generate a bounding box around an object, a label for the object, and a confidence score. An action is performed (240) responsive to the object and the confidence score.

The application discloses a sample data acquisition method and device, equipment and a storage medium, and belongs to the field of machine learning. In the embodiment of the application, in response to a sample acquisition request, an object identifier of a sample object and a content identifier of sample content recommended for the sample object are acquired, the sample content being a label of the sample object when a training content recommendation model is trained. Based on the object identifier, an object feature of the sample object is acquired, the object feature including an object behavior and an object label. In the case that the sample content is recommended by a target recommendation engine for the sample object, based on the content identifier, a content feature of the sample content is acquired from the target recommendation engine, the content feature being a content feature used by the target recommendation engine when the sample content is recommended for the sample object, thereby improving the consistency of the object feature and the content feature, and avoiding the problem that the content recommendation effect of the trained content recommendation model is poor due to feature crossing.

Embodiments of the present application disclose an object feature construction method and device and a computer readable storage medium, which can be applied to various scenes such as cloud technology, artificial intelligence, intelligent transportation and auxiliary driving. The object label information of a to-be-constructed object in a target system is obtained. The label similarity between object labels is calculated, and the basic object label is determined in the object label information based on the label similarity and the label weight. Semantic propagation is performed in a preset label relationship graph based on the semantic information of the basic object label, and the associated object label and the label association path are obtained. The number of nodes of the label association path is counted, and the label weight of the basic object label corresponding to the associated object label is adjusted according to the number of nodes to obtain the associated label weight. The object feature corresponding to the to-be-constructed object is constructed based on the associated object label and the associated label weight. In this way, the accuracy of object feature construction is improved, and the efficiency of object feature construction is improved.