Dual-module neural network structure video object segmentation method

Abstract

The invention provides a dual-module neural network structure video object segmentation method, for solving the problem that a video object segmentation result is not ideal due to noise interference in a video object segmentation process. The dual-module neural network structure video object segmentation method comprises the following steps: inputting a first frame image and a mask of a first frame into a transformation network to generate an image pair; generating a target proposal box for each image pair to determine whether the image pair is a region of interest; adding a tracker into the region of interest, inputting the tracker into a segmentation network of interest to train a learning model, and outputting the learning model; outputting a feature map from the last layer of convolution of the interested segmentation network and respectively inputting into a space attention module and a channel attention module; and finally, fusing the feature maps output by the two attention modules, and outputting a final segmentation mask result through convolution layer operation. According to the dual-module neural network structure video object segmentation method, a good segmentation experiment result is obtained on the DAVIS video data set.

Description

technical field [0001] The invention is in the field of computer vision, especially relates to video object segmentation processing with large-scale changes in video and inaccurate dynamic appearance changes, and specifically a method for video object segmentation with a dual-module neural network structure. Background technique [0002] In recent years, with the rapid development of computer vision technology, convolutional neural network in deep learning has received great attention in various research fields, and video object segmentation technology has become an important content that researchers have paid attention to in recent years. Video segmentation technology is increasingly showing its important position. Its applications in scene understanding, video labeling, driverless cars and object detection have all been rapidly developed in video segmentation technology. It can be said that the advancement of video segmentation technology drives the overall development of...

AI Technical Summary

Problems solved by technology

These methods usually use the first frame for data enhancement, and model adaptation relies heavily on the fine-tuning model. In the video, complex backgrounds, occlusions or fast movements, and camera shake oscillations cannot achieve efficient segmentation problems.

Images