Abstract: Target tracking methods and apparatuses, electronic devices, and storage media are provided. The method includes: obtaining features of a plurality of reference images of a target image; determining a plurality of initial predicted positions of a tracking target in the target image based on the features of the plurality of reference images; and determining a final position of the tracking target in the target image based on the plurality of initial predicted positions.

Abstract: Vehicle-mounted camera pose estimation methods, apparatuses, and systems, and electronic devices involve performing lane line detection of a road on which a vehicle drives on the basis of a video stream of the road acquired by a vehicle-mounted camera; obtaining horizon information of the road on which the vehicle drives according to a lane line detection result; and obtaining pose information of the vehicle-mounted camera according to the horizon information.

Abstract: Image detection methods and apparatuses, electronic devices and storage media include: down-sampling a second image to obtain a first image; determining at least one key sub-region in the second image for category detection; respectively extracting a feature of the first image and a feature of the at least one key sub-region; and generating a category detection result of the second image at least based on the extracted features.

Abstract: A method for identifying a target, a non-transitory computer-readable storage medium, and an electronic device include: acquiring a first image and a second image, the first image and the second image each including a target to be determined; generating a prediction path based on the first image and the second image, both ends of the prediction path respectively corresponding to the first image and the second image; and performing validity determination on the prediction path and determining, according to a determination result, whether the targets to be determined in the first image and the second image are the same target to be determined.

Abstract: A target object recognition method includes: performing target object detection on an object of an image to be detected to obtain target object prediction information of the object, where the target object prediction information is confidence information that the detected object is the target object; performing key point detection on the object of the image to be detected to obtain key point prediction information of the object, where the key point prediction information is confidence information that a key point of the detected object is a key point of the target object; fusing the target object prediction information with the key point prediction information to obtain comprehensive prediction information of the object; and recognizing the target object according to the comprehensive prediction information.

Abstract: A method for implementing image enhancement includes: performing filtering processing on a to-be-processed image to obtain an image subjected to the filtering processing; determining similarity degrees between pixel points in the to-be-processed image and a target region of a target object in the to-be-processed image; and fusing the similarity degrees, the to-be-processed image and the image subjected to the filtering processing, so that the higher a similarity degree between a pixel point and the target object in the to-be-processed image, the stronger a filtering effect of the pixel point, and the lower a similarity degree between the pixel point and the target object in the to-be-processed image, the weaker a filtering effect of the pixel point.

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: A gesture identification method includes: performing gesture information detection on an image by means of a neural network, to obtain a potential hand region, a potential gesture category and a potential gesture category probability in the image, the potential gesture category including a gesture-free category and at least one gesture category; and if the obtained potential gesture category with the maximum probability is the gesture-free category, not outputting position information of the potential hand region of the image; or otherwise, outputting the position information of the potential hand region of the image and the obtained potential gesture category with the maximum probability.

Abstract: A key frame scheduling method and apparatus include: performing feature extraction on a current frame through a first network layer of a neural network to obtain low-layer features of the current frame acquiring a scheduling probability of the current frame according to low-level features of a previous key frame adjacent to the current frame and the low-level features of the current frame; determining whether the current frame is scheduled as a key frame according to the scheduling probability value of the current frame; and when determining that the current frame is scheduled as a key frame, performing feature extraction on low-level features of a current key frame via a second network layer of the neural network to obtain high-level features of the current key frame, where the network depth of the first network layer is less than the network depth of the second network layer.

Abstract: The present disclosure relates to a neural network training method and apparatus, and an image processing method and apparatus. The training method includes: performing classification processing on a target image in a training set by means of a neural network to obtain a prediction classification result of the target image; and training the neural network according to the prediction classification result, and an initial category tag and a corrected category tag of the target image. Embodiments of the present disclosure may supervise the training process of the neural network by means of the initial category tag and the corrected category tag, and simplify the training process and a network structure.

Abstract: A search method, a search device, a storage medium and a computer program. The search method includes: determining a first similarity between text and at least one video, the text being used for representing a search condition; determining a first character interaction graph of the text and a second character interaction graph of the at least one video; determining a second similarity between the first character interaction graph and the second character interaction graph; and according to the first similarity and the second similarity, determining a video matching the search condition from the at least one video.

Abstract: A target detection method and apparatus, a target detection device and a storage medium. The target detection method comprises: acquiring n types of detection frames which are determined by n different target detectors and which are related to a positioning target, wherein the n different target detectors may extract n different types of image features from the same image, and n is a positive integer equal to or greater than 2; obtaining an integrated detection frame on the basis of the n types of detection frames; each target detector extracting image features from an input image on the basis of the integrated detection frame to obtain n types of image features; performing image feature optimization on the basis of the n types of image features to obtain optimized image features; and obtaining an optimized detection frame according to the optimized image features so as to detect a target.

Abstract: The present disclosure relates to an image processing method and apparatus, and a storage medium. The method includes: performing step-by-step convolution processing on an image to be processed to obtain a convolution result (S11); obtaining a positioning result through positioning processing according to the convolution result (S12); performing step-by-step deconvolution processing on the positioning result to obtain a deconvolution result (S13); and performing segmentation processing on the deconvolution result to segment a target object from the image to be processed (S14). Embodiments of the present disclosure implement target object positioning and segmentation at the same time in a process of image processing, and the image processing precision is improved while the speed of image processing is guaranteed.

Abstract: An image disambiguation method includes: performing image feature extraction and semantic recognition on at least two images in an image set including similar targets to obtain N K-dimensional semantic feature probability vectors, where the image set includes N images, N and K are both positive integers, and N is greater than or equal to 2; determining a differential feature combination according to the N K-dimensional semantic feature probability vectors, the differential feature combination indicating a difference between the similar targets in the at least two images in the image set; and generating a natural language for representing or prompting the difference between the similar targets in the at least two images in the image set according to the differential feature combination and image features of the at least two images in the image set.

Abstract: Methods and apparatuses for face image deduplication as well as non-transitory computer-readable storage medium include: filtering a plurality of obtained first face images to obtain at least one second face image with image quality reaching a first preset condition; matching the second face image with at least one third face image in an image queue to obtain a matching result; and determining, according to the matching result, whether to perform deduplication operation on the second face image.

Abstract: A method for image restoration, an electronic device and a computer storage medium are provided. The method includes that: region division is performed on an acquired image to obtain more than one sub-image; each sub-image is input into multiple paths of neural network and restored by using a restoration network determined for each sub-image; a restored image of each sub-image is output and obtained, so as to obtain a restored image of the acquired image.

Abstract: The present disclosure relates to a method and device for video processing, an electronic device, and a storage medium. The method comprises: determining, on the basis of paragraph information of a query text paragraph and video information of multiple videos in a video library, preselected videos associated with the query text paragraph in the multiple videos; and determining a target video in the preselected videos on the basis of video frame information of the preselected videos and of sentence information of the query text paragraph. The method for video processing of the embodiments of the present disclosure indexes videos by means of the relevance between the videos and the query text paragraph, allows the pinpointing of the target video, avoids search result redundancy, allows the processing of the query text paragraph in a natural language form, and is not limited by the inherent contents of content labels.

Abstract: A method for determining hand three-dimensional data includes: obtaining a first hand image and a second hand image captured by a binocular photographing system; identifying, from each of the first hand image and the second hand image, at least one key point and a region profile covering the at least one key point; determining depth information of the at least one key point and depth information of the region profile according to a photographing parameter of the binocular photographing system, the at least one key point and the region profile identified from the first hand image, and the at least one key point and the region profile identified from the second hand image; and determining hand three-dimensional data according to the at least one key point and the depth information of the at least one key point together with the region profile and the depth information of the region profile.

Abstract: Body contour key point detection methods, image processing methods, neural network training methods, apparatuses, electronic devices, computer-readable storage media, and computer programs include: obtaining an image feature of an image block including a body; obtaining a body contour key point prediction result of the body by means of a first neural network according to the image feature; and obtaining a body contour key point in the image block according to the body contour key point prediction result; where the body contour key point is used for representing an outer contour of the body.

Abstract: An orientation detection method and device, electronic device and storage medium, the method comprising: determining first position information of at least one first feature part of a target object in a target image (S100); determining three-dimensional position information of a second feature part of the target object on the basis of the first position information and device parameters of a camera device (S200); and determining spatial orientation of the target object on the basis of the first position information of the at least one first feature part comprised in the second feature part and the three-dimensional position information of the second feature part (S300). The described method may increase the accuracy of orientation detection.