Abstract: This application relates to the field of artificial intelligence and the field of computer vision. The method includes performing feature extraction on an image to obtain a basic feature map of the image, and determining a proposal of a region possibly including a pedestrian in the image. The basic feature map of the image is then processed to obtain an object visibility map in which a response to a pedestrian visible part is greater than a response to a pedestrian blocked part and a background part. The method further performs weighted summation processing on the object visibility map and the basic feature map to obtain an enhanced feature map of the image, and determines, based on the proposal of the image and the enhanced feature map of the image, a bounding box including a pedestrian in the image and a confidence level of the bounding box including the pedestrian.

Abstract: A method for optimizing a photographing pose of a user, where the method is applied to an electronic device, and the method includes: displaying a photographing interface of a camera of the electronic device; obtaining a to-be-taken image in the photographing interface; determining, based on the to-be-taken image, that the photographing interface includes a portrait; entering a pose recommendation mode; and presenting a recommended human pose picture to a user in a predetermined preview manner, where the human pose picture is at least one picture that is selected from a picture library through metric learning and that has a top-ranked similarity to the to-be-taken image, and where the similarity is an overall similarity obtained by fusing a background similarity and a foreground similarity.

Abstract: A method for optimizing a photographing pose of a user, where the method is applied to an electronic device, and the method includes: displaying a photographing interface of a camera of the electronic device; obtaining a to-be-taken image in the photographing interface; determining, based on the to-be-taken image, that the photographing interface includes a portrait; entering a pose recommendation mode; and presenting a recommended human pose picture to a user in a predetermined preview manner, where the human pose picture is at least one picture that is selected from a picture library through metric learning and that has a top-ranked similarity to the to-be-taken image, and where the similarity is an overall similarity obtained by fusing a background similarity and a foreground similarity.

Abstract: An image processing method. The method includes: An electronic device obtains N images, where the N images have a same quantity of pixels and a same pixel location arrangement, and N is an integer greater than 1; the electronic device obtains, based on feature values of pixels located at a same location in the N images, a reference value of the corresponding location; the electronic device determines a target pixel of each location based on a reference value of the location; and the electronic device generates a target image based on the target pixel of each location.

Abstract: This application discloses an image segmentation method in the field of artificial intelligence. The method includes: obtaining an input image and a processing requirement; performing multi-layer feature extraction on the input image to obtain a plurality of feature maps; downsampling the plurality of feature maps to obtain a plurality of feature maps with a reference resolution, where the reference resolution is less than a resolution of the input image; fusing the plurality of feature maps with the reference resolution to obtain at least one feature map group; upsampling the feature map group by using a transformation matrix W, to obtain a target feature map group; and performing target processing on the target feature map group based on the processing requirement to obtain a target image.

Abstract: An image processing method includes detecting human skeleton key points of a character in an image, and for the to-be-processed image, generating a cropping region based on the human skeleton key points and a preset rule, and using the to-be-processed image selected by using the cropping region as an output object. The human skeleton key points and the preset rule are set based on photographing experience.

Abstract: This application provides a model training method in the artificial intelligence field. In a process of determining a loss used to update a model parameter, factors are comprehensively considered. Therefore, an obtained neural network has a strong generalization capability. The method in this application includes: obtaining a first source domain image associated with a target domain image and a second source domain image associated with the target domain image; obtaining a first prediction label of the first source domain image and a second prediction label of the second source domain image through a first to-be-trained model; obtaining a first loss based on the first prediction label and the second prediction label, where the first loss indicates a difference between the first prediction label and the second prediction label; and updating a parameter of the first to-be-trained model based on the first loss, to obtain a first neural network.

Abstract: In a method for selecting pictures from a sequence of pictures of an object in motion, a computerized user device determines, for each picture in the sequence of pictures, a value of a motion feature of the object. Based on analyzing the values of the motion feature of the pictures in the sequence, the device identifies a first subset of pictures from the pictures in the sequence. The device then selects, based on a second selection criterion, a second subset of pictures from the first subset of pictures. The pictures in the second subset are displayed to a user for further selection.

Abstract: An action recognition method and apparatus related to artificial intelligence and include extracting a spatial feature of a to-be-processed picture, determining a virtual optical flow feature of the to-be-processed picture based on the spatial feature and X spatial features and X optical flow features in a preset feature library, where the X spatial features and the X optical flow features include a one-to-one correspondence, determining a first type of confidence of the to-be-processed picture in different action categories based on similarities between the virtual optical flow feature and Y optical flow features, where each of the Y optical flow features in the preset feature library corresponds to one action category, X and Y are both integers greater than 1, and determining an action category of the to-be-processed picture based on the first type of confidence.

Abstract: This application relates to Artificial intelligence and provides a method for training a classifier, one example method including: obtaining a first training sample, where the first training sample includes a corresponding semantic tag; obtaining a plurality of second training samples, where each of the second training samples includes a corresponding semantic tag; determining a target sample from the plurality of second training samples based on semantic similarities between the first training sample and the plurality of second training samples; and training the classifier based on the first training sample, the target sample, and a semantic similarity between the first training sample and the target sample.

Abstract: An action recognition method and apparatus related to artificial intelligence and include extracting a spatial feature of a to-be-processed picture, determining a virtual optical flow feature of the to-be-processed picture based on the spatial feature and X spatial features and X optical flow features in a preset feature library, where the X spatial features and the X optical flow features include a one-to-one correspondence, determining a first type of confidence of the to-be-processed picture in different action categories based on similarities between the virtual optical flow feature and Y optical flow features, where each of the Y optical flow features in the preset feature library corresponds to one action category, X and Y are both integers greater than 1, and determining an action category of the to-be-processed picture based on the first type of confidence.

Abstract: Action recognition methods are disclosed.

Abstract: A video clip location technology in the field of computer vision pertaining to artificial intelligence that provides a video processing method and apparatus. The method includes: obtaining a semantic feature of an input sentence; performing semantic enhancement on a video frame based on the semantic feature to obtain a video feature of the video frame, where the video feature includes the semantic feature; and determining, based on the semantic feature and the video feature, whether a video clip to which the video frame belongs is a target video clip corresponding to the input sentence. The method helps improve accuracy of recognizing a target video clip corresponding to an input sentence.

Abstract: A method for optimizing a photographing pose of a user, where the method is applied to an electronic device, and the method includes: displaying a photographing interface of a camera of the electronic device; obtaining a to-be-taken image in the photographing interface; determining, based on the to-be-taken image, that the photographing interface includes a portrait; entering a pose recommendation mode; and presenting a recommended human pose picture to a user in a predetermined preview manner, where the human pose picture is at least one picture that is selected from a picture library through metric learning and that has a top-ranked similarity to the to-be-taken image, and where the similarity is an overall similarity obtained by fusing a background similarity and a foreground similarity.

Abstract: An action recognition method and apparatus related to artificial intelligence and include extracting a spatial feature of a to-be-processed picture, determining a virtual optical flow feature of the to-be-processed picture based on the spatial feature and X spatial features and X optical flow features in a preset feature library, where the X spatial features and the X optical flow features include a one-to-one correspondence, determining a first type of confidence of the to-be-processed picture in different action categories based on similarities between the virtual optical flow feature and Y optical flow features, where each of the Y optical flow features in the preset feature library corresponds to one action category, X and Y are both integers greater than 1, and determining an action category of the to-be-processed picture based on the first type of confidence.

Abstract: This application relates to the field of artificial intelligence, and specifically, to the field of computer vision. The method includes: performing feature extraction on an image to obtain a basic feature map of the image; determining a proposal of a region possibly including a pedestrian in the image; processing the basic feature map of the image to obtain an object visibility map in which a response to a pedestrian visible part is greater than a response to a pedestrian blocked part and a background part; performing weighted summation processing on the object visibility map and the basic feature map to obtain an enhanced feature map of the image; and determining, based on the proposal of the image and the enhanced feature map of the image, a bounding box including a pedestrian in the image and a confidence level of the bounding box including the pedestrian in the image.

Abstract: In one embodiment, a target tracking method includes: receiving a current frame of picture including a target object; determining, based on a drift determining model, whether a tracker drifts for tracking of the target object in the current frame of picture, where the drift determining model is obtained through modeling based on largest values of responses values of a training sample used to train the drift determining model, where the training sample is collected from a training picture that includes the target object, where the response value of the sample is a value indicating a probability that the training sample is the target object in the training picture; and outputting a tracking drift result, where the tracking drift result includes: drift is generated for the tracking of the target object, or no drift is generated for the tracking of the target object.

Abstract: This application discloses an object recognition method and apparatus in the field of artificial intelligence. This application relates to the field of artificial intelligence, and specifically, to the field of computer vision. The method includes: obtaining one or more body regions of a to-be-recognized image; determining a saliency score of each of the one or more body regions; and when a saliency score of a body region A is greater than or equal to a categorization threshold, determining a feature vector of an object in the body region A based on a feature of the object in the body region A, and determining a category of the object in the body region A based on the feature vector of the object in the body region A and a category feature vector in a feature library, where the body region A is any one of the one or more body regions.

Abstract: An image processing method. The method includes: An electronic device obtains N images, where the N images have a same quantity of pixels and a same pixel location arrangement, and N is an integer greater than 1; the electronic device obtains, based on feature values of pixels located at a same location in the N images, a reference value of the corresponding location; the electronic device determines a target pixel of each location based on a reference value of the location; and the electronic device generates a target image based on the target pixel of each location.

Abstract: This application discloses an image segmentation method in the field of artificial intelligence. The method includes: obtaining an input image and a processing requirement; performing multi-layer feature extraction on the input image to obtain a plurality of feature maps; downsampling the plurality of feature maps to obtain a plurality of feature maps with a reference resolution, where the reference resolution is less than a resolution of the input image; fusing the plurality of feature maps with the reference resolution to obtain at least one feature map group; upsampling the feature map group by using a transformation matrix W, to obtain a target feature map group; and performing target processing on the target feature map group based on the processing requirement to obtain a target image.