Optimized matting method and system capable of specifying category

Abstract

The invention discloses an optimized matting method and system capable of specifying a category. A semantic segmentation network of multi-class prediction is adopted to generate the ternary images ofdifferent class foregrounds of a to-be-tested image, a certain class of ternary images in the image can be rapidly obtained through the connected domain calculation, then an image matting network is adopted to carry out the refined image matting according to the selected class of ternary images and the to-be-tested image, so that an alpha image with optimized quality is obtained. According to thepresent invention, the foreground part of the specified category in the image can be automatically extracted, a transparent base map meeting various scene requirements is generated, the later image processing efficiency can be improved, the marketing personnel can also be helped to generate the marketing data, such as posters, etc., the commodity conversion rate is effectively improved, and the labor cost is reduced.

Description

technical field [0001] The present invention relates to a technology in the field of artificial intelligence image processing, in particular to a method and system for optimizing image matting that can specify a category. Background technique [0002] Cutout technology has great application prospects and user needs in e-commerce and other scenarios. The traditional matting technology is often based on the ternary image (including the determined foreground part, the determined background part and the uncertain part combined with the foreground background) according to the color, texture and other low-level features of the picture to infer the uncertain pixels from the known determined pixel part part, without using advanced semantic information, this method is not good for images with similar foreground and background colors or complex textures, and it takes a long time to process pixel by pixel. It can be seen that the traditional matting method is difficult to meet Daily a...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems of low quality, poor matting effect and long time required for existing automatically generated ternary images, the present invention proposes an optimized matting method and system that can specify a category, which can automatically extract the foreground of a specified category in an image part, generate a transparent base map (alpha map) that meets the requirements of various scenarios, which can help improve the efficiency of post-image processing, and can also be used by marketing personnel to generate marketing materials such as posters, effectively improving product conversion rates and reducing labor costs

Images