Abstract: A method of establishing an image set for image recognition includes obtaining a single-label image set comprising an image annotated with a single label, and a multi-label image set comprising an image annotated with a plurality of labels; converting content of each label into a corresponding word identifier according to a semantic network, to obtain a word identifier set, a converted single-label image set, and a converted multi-label image set; and constructing a hierarchical semantic structure according to the word identifier set and the semantic network. The method also includes performing label supplementation on the image in the converted single-label image set to obtain a supplemented single-label image set; performing label supplementation on the supplemented single-label image set to obtain a final supplemented image set; and establishing a target multi-label image set to train an image recognition model by using the target multi-label image set.

Abstract: A computer device obtains a first face image (IMA) and a second face image (IFA). The device obtains M first image blocks corresponding to facial features from the first face image (IMA), and obtains N second image blocks corresponding to facial features from the second face image (IFA). The device transforms the M first image blocks and the N second image blocks to a feature space to generate M first feature blocks and N second feature blocks. The device selects a subset of the first feature blocks and a subset of the second feature blocks according to a specified control vector. The device generates a first composite feature map based the selected subset of the first feature blocks and the selected subset of the second feature blocks. The device inversely transforms the first composite feature map back to an image space to generate a third face image.

Abstract: The present disclosure describes a method, apparatus, and storage medium for annotating image. The method includes extracting, by a device, a visual feature of an image through a generative adversarial network model, and sequentially inputting M pieces of random noise into the generative adversarial network model. In response to each of the M pieces of random noise being inputted into the generative adversarial network model, the method includes performing a determinantal point process (DPP) on the visual feature of the image and the each random noise through the generative adversarial network model to obtain N tag subsets, and selecting a distinct tag subset from the N tag subsets through the generative adversarial network model. The method also includes outputting M distinct tag subsets through the generative adversarial network model after the M pieces of random noise are inputted into the generative adversarial network model.

Abstract: An image processing method includes: obtaining a target image; performing feature extraction on the target image based on a residual network, to obtain image feature information; and performing recognition processing on the target image according to the image feature information. The residual network includes a plurality of residual blocks that are successively connected, each of the residual blocks including a convolution branch and a residual branch, a size of a convolution kernel of a first convolutional layer in the convolution branch being less than a size of a convolution kernel of a second convolutional layer located after the first convolutional layer, and a convolution stride of the second convolutional layer being greater than a convolution stride of the first convolutional layer and less than a width of the convolution kernel of the second convolutional layer.

Abstract: Aspects of the disclosure are directed to a training method and apparatus of an adversarial attack model, a generating method and apparatus of an adversarial image, an electronic device, and a storage medium. The adversarial attack model can include a generator network, and the training method can include using the generator network to generate an adversarial attack image based on a training digital image, and performing an adversarial attack on a target model based on the adversarial attack image, to obtain an adversarial attack result. The training method can further include obtaining a physical image corresponding to the training digital image, and training the generator network based on the training digital image, the adversarial attack image, the adversarial attack result, and the physical image.

Abstract: A facial expression recognition method includes extracting a first feature from color information of pixels in a first image, and extracting a second feature of facial key points from the first image.

Abstract: A computer device obtains a first face image (IMA) and a second face image (IFA). The device obtains M first image blocks corresponding to facial features from the first face image (IMA), and obtains N second image blocks corresponding to facial features from the second face image (IFA). The device transforms the M first image blocks and the N second image blocks to a feature space to generate M first feature blocks and N second feature blocks. The device selects a subset of the first feature blocks and a subset of the second feature blocks according to a specified control vector. The device generates a first composite feature map based the selected subset of the first feature blocks and the selected subset of the second feature blocks. The device inversely transforms the first composite feature map back to an image space to generate a third face image.

Abstract: A model training method and apparatus for image recognition, and a non-transitory storage medium are provided. The model training method includes: obtaining a multi-label image training set including a plurality of training images each annotated with a plurality of sample labels; selecting target training images from the multi-label image training set for training a current model; performing label prediction on each target training image using the current model, to obtain a plurality of predicted labels of the each target training image; obtaining a cross-entropy loss function corresponding to the plurality of sample labels of the each target training image, a positive label loss being greater than a negative label loss and having a weight greater than 1; converging the predicted labels and the sample labels of the each target training image according to the cross-entropy loss function, and updating parameters of the current model, to obtain a trained model.

Abstract: A method of establishing an image set for image recognition includes obtaining a single-label image set comprising an image annotated with a single label, and a multi-label image set comprising an image annotated with a plurality of labels; converting content of each label into a corresponding word identifier according to a semantic network, to obtain a word identifier set, a converted single-label image set, and a converted multi-label image set; and constructing a hierarchical semantic structure according to the word identifier set and the semantic network. The method also includes performing label supplementation on the image in the converted single-label image set to obtain a supplemented single-label image set; performing label supplementation on the supplemented single-label image set to obtain a final supplemented image set; and establishing a target multi-label image set to train an image recognition model by using the target multi-label image set.

Abstract: An image processing method includes: obtaining a target image; performing feature extraction on the target image based on a residual network, to obtain image feature information; and performing recognition processing on the target image according to the image feature information. The residual network includes a plurality of residual blocks that are successively connected, each of the residual blocks including a convolution branch and a residual branch, a size of a convolution kernel of a first convolutional layer in the convolution branch being less than a size of a convolution kernel of a second convolutional layer located after the first convolutional layer, and a convolution stride of the second convolutional layer being greater than a convolution stride of the first convolutional layer and less than a width of the convolution kernel of the second convolutional layer.

Abstract: The present disclosure describes a method, apparatus, and storage medium for annotating image. The method includes extracting, by a device, a visual feature of an image through a generative adversarial network model, and sequentially inputting M pieces of random noise into the generative adversarial network model. In response to each of the M pieces of random noise being inputted into the generative adversarial network model, the method includes performing a determinantal point process (DPP) on the visual feature of the image and the each random noise through the generative adversarial network model to obtain N tag subsets, and selecting a distinct tag subset from the N tag subsets through the generative adversarial network model. The method also includes outputting M distinct tag subsets through the generative adversarial network model after the M pieces of random noise are inputted into the generative adversarial network model.