Abstract: A virtual avatar generation method includes: determining a target task associated with at least one target face property, where the at least one target face property is one of a plurality of predefined face properties respectively; performing, according to the target task, target face property analysis on a target image including at least a face to obtain a target face property feature associated with the target face property of the target image; determining a target virtual avatar template corresponding to the target face property feature according to predefined correspondence between face property features and virtual avatar templates; and generating a virtual avatar of the target image based on the target virtual avatar template.

Abstract: The embodiments disclose a method and device for liveness detection and a computer-readable storage medium. The method includes that: a first image collected by a first image sensor of a binocular camera is acquired, and a second image collected by a second image sensor of the binocular camera is acquired: binocular matching processing is performed on the first image and the second image to obtain parallax information of the first image and the second image; and a Iiveness detection result is obtained according to the parallax information. By adopting the embodiments of the present disclosure, the accuracy of liveness detection can be improved.

Abstract: An image processing method and apparatus, an electronic device, and a storage medium are provided. The method includes: obtaining an infrared image and a color image for the same object in a predetermined scenario, and a first depth map corresponding to the color image; obtaining a first optical flow between the color image and the infrared image; and performing first optimization processing on the first depth map by using the first optical flow to obtain an optimized second depth map corresponding to the color image.

Abstract: Methods and apparatuses for deep learning-based recommendation, electronic devices, and media include: respectively obtaining related information of a target user and related information of a target item; respectively using at least two sub-models in an integrated model to obtain, based on the related information of the target user and the related information of the target item, operating probabilities corresponding to the at least two sub-models; obtaining, based on the operating probabilities corresponding to the at least two sub-models, a target probability about the target user operating the target item; and recommending the target item to the target user based on the target probability.

Abstract: An authentication method includes: information to be verified is encrypted according to a first time stamp to obtain a first ciphertext; an authentication request carrying the first ciphertext and the first time stamp is sent to a server; an authentication response corresponding to the authentication request and sent by the server is received, the authentication response carrying a second ciphertext and a second time stamp; and the second ciphertext is decrypted according to the second time stamp to obtain an authentication result of authenticating the information to be verified by the server.

Abstract: A method for generating a special effect program file package includes: importing a group of sub-materials, where the group of sub-materials include a plurality of sub-materials; obtaining parameter values of playback parameters of the group of sub-materials; and generating a special effect program file package according to the group of sub-materials and the parameter values of the playback parameters.

Abstract: Method and apparatus for generating a road map, electronic device, and non-transitory computer storage medium are disclosed, including: inputting a remote sensing image into a first neural network to extract first road feature information of multiple channels via the first neural network; inputting the first road feature information of multiple channels into a third neural network, to extract third road feature information of multiple channels via the third neural network, where the third neural network is a neural network trained by using road direction information as supervision information; fusing the first road feature information and the third road feature information; and generating a road map according to a fusion result.

Abstract: Embodiments of the present application disclose methods and apparatuses for determining a bounding box of a target object, media, and devices. The method includes: obtaining attribute information of each of a plurality of key points of a target object; and determining a bounding box position of the target object according to the attribute information of each of the plurality of key points of the target object and to a preset neural network. The implementations of the present application can improve the efficiency and accuracy of determining a bounding box of a target object.

Abstract: A face image processing method and apparatus, an electronic device, and a storage medium are provided. The method includes: obtaining face key points and a face deflection angle in a face image; determining a submalar triangle center in the face image according to the face key points and the face deflection angle; determining a submalar triangle region in the face image according to the face key points and the submalar triangle center; and performing color filling on the submalar triangle region. By means of the present disclosure, a submalar triangle region can be accurately positioned, and submalar triangle filling is performed based on the accurately positioned submalar triangle region, thereby obtaining a more natural filling effect.

Abstract: A face posture analysis method, an electronic device, and a computer-readable storage medium are provided. The face posture analysis method includes: obtaining a face key point of a to-be-processed face image; and inputting the face key point of the to-be-processed face image into a neural network, to obtain face posture information of the to-be-processed face image output by the neural network.

Abstract: Examples of the present disclosure disclose data processing methods, apparatuses and storage media. The data processing method includes: obtaining a plurality of video images of a target site within a predetermined time range; performing pedestrian identification processing on the plurality of video images to obtain, based on the identification result, a customer counting result for a first region in the target site and a customer counting result for a second region in the target site; and determining a degree of association between the second region and the first region based on the customer counting result for the first region and the customer counting result for the second region.

Abstract: An image processing method includes: inputting a to-be-processed image into a neural network; and forming discrete feature data of the to-be-processed image via the neural network, where the neural network is trained based on guidance information, and during the training process, the neural network is taken as a student neural network; the guidance information includes: a difference between discrete feature data formed by a teacher neural network for an image sample and discrete feature data formed by the student neural network for the image sample.

Abstract: A method for liveness detection includes: acquiring a first depth map captured by a depth sensor and a first target image captured by an image sensor; performing quality detection on the first depth map to obtain a quality detection result of the first depth map; and determining a liveness detection result of a target object in the first target image based on the quality detection result of the first depth map. The present disclosure can improve the accuracy of liveness detection.

Abstract: An object detection method, a neural network training method, an apparatus, and an electronic device include: obtaining, through prediction, multiple fused feature graphs from images to be processed, through a deep convolution neural network for target region frame detection, obtaining multiple first feature graphs from a first subnet having at least one lower sampling layer, obtaining multiple second feature graphs from a second subnet having at least one upper sampling layer, and obtaining fused graph by fusing multiple first feature graphs and multiple second feature graphs respectively; and obtaining target region frame data according to the multiple fused feature graphs. Because the fused feature graphs better represent semantic features on high levels and detail features on low levels in images, target region frame data of big and small objects in images can be effectively extracted according to the fused feature graphs, thereby improving accuracy and robustness of object detection.

Abstract: Data processing methods, data processing apparatuses, and storage media are provided. The method is applicable to a data processing system. The data processing system includes a storage device and a programmable device. Data is transmitted between the storage device and the programmable device via a bus. A controller and an accelerator are deployed on the programmable device. The controller is enabled with at least two kinds of data format conversion function. The method includes: the controller obtaining the first data; the controller performing data format conversion on the first data to obtain second target data in a target data format; and the controller storing the second data to the storage device and/or sending the second data to the accelerator.

Abstract: Provided are methods and apparatuses for positioning face feature points. The method includes: carrying out edge detection on a face image to obtain a face feature line image; and fusing the face image and the face feature line image to obtain position information of face feature points.

Abstract: The present disclosure relates to a network training method and device, an image processing method and device, the method comprising: performing pixel shuffling on a first image in a training set to obtain a second image, wherein the first image is an image subjected to pixel shuffling; performing, by a feature extraction network of a neural network, feature extraction on the first image to obtain a first image feature, and performing, by a feature extraction network, feature extraction on the second image to obtain a second image feature; performing, by a recognition network of the neural network, recognition on the first image feature to obtain a recognition result of the first image; and training the neural network according to the recognition result, the first image feature and the second image feature. Embodiments of the present disclosure enable improvement of recognition precision of neural networks.

Abstract: An anchor determination includes: performing feature extraction on an image to be processed to obtain a first feature map of the image to be processed; performing anchor prediction on the first feature map via an anchor prediction network to obtain position information of anchors and shape information of the anchors in the first feature map, the position information of the anchors referring to information about positions in the first feature map where the anchors are generated. A corresponding anchor determination apparatus and a storage medium are also provided.

Abstract: A semantic segmentation model training method includes: performing, by a semantic segmentation model, image semantic segmentation on at least one unlabeled image to obtain a preliminary semantic segmentation result as the category of the unlabeled image; obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and the category of at least one labeled image, sub-images respectively corresponding to the at least two images and features corresponding to the sub-images, where the at least two images comprise the at least one unlabeled image and the at least one labeled image, and the at least two sub-images carry the categories of the corresponding images; and training the semantic segmentation model on the basis of the categories of the at least two sub-images and feature distances between the at least two sub-images.

Abstract: A pedestrian re-identification method includes: obtaining a target video containing a target pedestrian and at least one candidate video; encoding each target video segment in the target video and each candidate video segment in the at least one candidate segment separately; determining a score of similarity between the each target video segment and the each candidate video segment according to encoding results, the score of similarity being used for representing a degree of similarity between pedestrian features in the target video segment and the candidate video segment; and performing pedestrian re-identification on the at least one candidate video according to the score of similarity.