Image annotation method and device, image semantic segmentation method and device and model training method and device

Abstract

The invention provides an image annotation method and device, an image semantic segmentation method and device, and a model training method and device, relates to the technical field of artificial intelligence, and is used for improving the sample image annotation efficiency. According to the image annotation method, the edge pixel points in the sample image are detected, the target image blocks in the image blocks in the sample image are screened according to the edge pixel points, and the target image blocks are annotated, so that the annotation result of the sample image is obtained, and due to the fact that all the pixel points in the sample image do not need to be annotated, the annotation quantity in the sample annotation process can be relatively reduced, and the efficiency of annotating the sample image is improved; and as the image has certain redundant information, the accuracy of the image semantic segmentation model is not influenced even if all pixel points in the sample image are not annotated and the image semantic segmentation model is trained.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to the field of artificial intelligence technology, and provides a method and device for image annotation, image semantic segmentation, and model training. Background technique [0002] At present, a variety of image segmentation models have gradually emerged, which are used to segment various types of images. The image segmentation model involves an important image segmentation model, that is, the image semantic segmentation model. The image semantic segmentation model generally classifies the image at the pixel level, and can finely segment the image. [0003] Most of the image semantic segmentation models are obtained by supervised learning, that is, a large number of sample images are required to train the image semantic segmentation model in order to obtain the trained image semantic segmentation model. The following takes two of the image semantic segmentation mo...

AI Technical Summary

Problems solved by technology

This semantic segmentation method can resist the learning loss function to express the training loss of the model, but the model still needs manual annotation results corresponding to the entire image during the training phase, and the process of annotating images will consume a lot of annotation costs

[0006] It can be seen that no matter which semantic segmentation method is used, the images required in the process of training the model need to be manually annotated pixel by pixel, and manually annotating a sample image will take at least 10 minutes, and the training process requires a large number of annotated images. images, which results in a less efficient labeling process

Images