Abstract: A method for training a multi-target image-text matching model and an image-text retrieval method are provided. The method for training the multi-target image-text matching model includes: obtaining a plurality of training samples that include sample pairs each including a sample image and a sample text, the sample image including a plurality of targets; obtaining, for each of the plurality of training samples, a heat map corresponding to the sample text in the training sample, the heat map representing a region of the target in the sample image that corresponds to the sample text; and training an image-text matching model based on a plurality of the sample texts and corresponding heat maps to obtain the multi-target image-text matching model.

Abstract: An object area measurement method and an apparatus are provided, relating to the computer vision and deep learning technology. The method includes acquiring an original image with a spatial resolution, the original image including a target object; acquiring an object identification model including at least two sets of classification models; generating one or more original image blocks based on the original image; performing operations on each original image block: scaling each original image block at at least two scaling levels to obtain scaled image blocks with at least two sizes, the scaled image blocks respectively corresponding to the at least two sets of classification models, and inputting the scaled image blocks into the object identification model to obtain an identification result of the target object; and determining an area of the target object based on the respective identification results of the one or more original image blocks and the spatial resolution.

Abstract: A method for recognizing a category of an image includes: acquiring a spectral image; training an image recognition model based on the spectral image, in which the image recognition model acquires a spectral semantic feature of each pixel, a minimum distance between each pixel and each category, and a spectral distance between a first spectrum of each pixel and a second spectrum of each category; splices them; and performs classification and recognition based on the spliced feature to output a recognition probability of each pixel under each category; determining a loss function of the image recognition model, adjusting the image recognition model based on the loss function, and returning to training the adjusted image recognition model based on the spectral image until training ends; recognizing a maximum recognition probability, output from a target image recognition model, and using a category corresponding to the maximum recognition probability as a target category.

Abstract: A training method for a human body attribute detection model includes: acquiring positive sample sub-images and negative sample sub-images respectively corresponding to a plurality of human body attribute categories; determining a plurality of first annotation attributes respectively corresponding to the plurality of positive sample sub-images; and a plurality of second annotation attributes respectively corresponding to the plurality of negative sample sub-images; and training an artificial intelligence model according to the plurality of positive sample sub-images, the plurality of negative sample sub-images, the plurality of first annotation attributes and the plurality of second annotation attributes to obtain the human body attribute detection model, so that the human body attribute detection model obtained by training can effectively model fine-grained attributes of the human body.

Abstract: A method for generating a road annotation, a device, and a storage medium are provided. The method may include: generating a road quantity and a road width in a tag picture; generating, for each road in the tag picture, a start point and an end point of the road; generating at least one point between the start point and the end point; drawing, for two adjacent points, a line segment from a previous point to a next point, where a width of the line segment is equal to the road width; and generating slanted box annotation information based on a coordinate of the previous point and a coordinate of the next point, where the slanted box annotation information includes an intersection point of diagonal lines, a width, a height and a slant angle of a slanted box.

Abstract: A method of training an image-text retrieval model, a method of multimodal image retrieval, an electronic device and a storage medium, each relating to the technical field of artificial intelligence, and in particular, to fields of computer vision and deep learning technologies. Sample data including a sample text and a sample image is acquired. The sample text includes a sample text in a first language and a sample text in a second language. The sample text in the first language and the sample text in the second language are processed by using the text encoding sub-model to obtain a sample text feature of the sample data. The sample image is processed by using the image encoding sub-model to obtain a sample image feature of the sample data. The image-text retrieval model is trained according to the sample text feature and the sample image feature.

Abstract: A method for generating a sample image includes: obtaining an initial image size of an initial image; obtaining a plurality of reference images by processing the initial image based on different reference processing modes; obtaining an image to be processed by fusing the plurality of reference images; and determining a target sample image from images to be processed based on the initial image size.

Abstract: Provided are a training method and apparatus for a target detection model, a device and a storage medium. The training method is described below. A feature map of a sample image is processed through a classification network of an initial model and a heat map and a classification prediction result of the feature map are obtained, a classification loss value is determined according to the classification prediction result and classification supervision data of the sample image, and a category probability of pixels in the feature map is determined according to the heat map of the feature map and a probability distribution map of the feature map is obtained; the feature map is processed through a regression network of the initial model and a regression prediction result is obtained, and a regression loss value is determined.

Abstract: The disclosure provides a method for detecting a change of a building, an apparatus for detecting a change of a building, an electronic device, a storage medium and a computer program product. The method includes: obtaining a remote-sensing image sequence of a to-be-detected region; obtaining a building probability map corresponding to each remote-sensing image in the remote-sensing image sequence; determining a sub-region located by each building in the to-be-detected region based on the building probability map corresponding to each remote-sensing image; for each building, determining an existence probability of the building in each remote-sensing image based on the sub-region located by the building and the building probability map corresponding to each remote-sensing image; and determining a change condition of the building based on the existence probability of the building in each remote-sensing image.

Abstract: The disclosure provides a method for recognizing an image, an apparatus for recognizing an image, an electronic device and a storage medium. An image to be processed is obtained. The number of first channels of the image is greater than the number of second channels of a red-green-blue (RGB) image. For each pixel of the image, a semantic type of the pixel is determined based on a value of the pixel on each channel. A recognition result of the image is generated based on the image and the semantic type of each pixel.

Abstract: An object detection model training method, object detection method and related apparatus, relate to the field of artificial intelligence technologies such as computer vision, deep learning. An implementation includes: obtaining training sample data including a first remote sensing image and position annotation information of an anchor box of a subject to be detected in the first remote sensing image, where the position annotation information includes angle information of the anchor box relative to a preset direction; obtaining an object feature map of the first remote sensing image based on an object detection model, performing object detection on the subject to be detected based on the object feature map to obtain an object bounding box, and determining loss information between the anchor box and the object bounding box based on the angle information; updating a parameter of the object detection model based on the loss information.

Abstract: An object area measurement method and an apparatus are provided, relating to the computer vision and deep learning technology. The method includes acquiring an original image with a spatial resolution, the original image including a target object; acquiring an object identification model including at least two sets of classification models; generating one or more original image blocks based on the original image; performing operations on each original image block: scaling each original image block at at least two scaling levels to obtain scaled image blocks with at least two sizes, the scaled image blocks respectively corresponding to the at least two sets of classification models, and inputting the scaled image blocks into the object identification model to obtain an identification result of the target object; and determining an area of the target object based on the respective identification results of the one or more original image blocks and the spatial resolution.

Abstract: The present disclosure discloses a method and apparatus for generating an image, a device, a storage medium and a program product, relates to the field of artificial intelligence, and particularly to computer vision and deep learning technologies, and may be applied in smart cloud and power grid inspection scenarios. A particular implementation of the method comprises: acquiring an original insulator image; performing an image transformation on the original insulator image to obtain a composite insulator image; and inputting the original insulator image and the composite insulator image into a pre-trained generative adversarial network to generate a target insulator image. According to the implementation, the image transformation is performed on the original insulator image, and then, massive target insulator images are generated through the generative adversarial network.

Abstract: The present disclosure discloses a method and apparatus for training a model, a method and apparatus for predicting a mineral, a device and a storage medium, and relates to the fields of computer vision and deep learning technologies. An implementation of the method may include: acquiring a target hyperspectral image of a target area, the target hyperspectral image including at least one pixel point annotated with a mineral category; determining a mask image corresponding to the target hyperspectral image; determining a sample hyperspectral image according to the target hyperspectral image and the mask image; determining an annotation vector of each pixel point according to the at least one pixel point annotated with the mineral category; and training a model according to the sample hyperspectral image and the annotation vector of the each pixel point.

Abstract: A header model for instance segmentation includes a target box branch having a first branch and a second branch, where the first branch is configured to process an inputted first feature map to obtain class information and confidence of a target box, and the second branch is configured to process the first feature map to obtain location information of the target box. The header model also includes a mask branch configured to process an inputted second feature map to obtain mask information, wherein the second feature map is a feature map outputted by an ROI extraction module, and the first feature map is a feature map resulting from a pooling performed on the second feature map.