AVS-试听分割-论文阅读

题目： Audio-Visual Segmentation

论文地址：https://arxiv.org/abs/2207.05042

GitHub地址：https://github.com/OpenNLPLab/AVSBench

项目主页：https://opennlplab.github.io/AVSBench/
相关博客https://arxiv.org/abs/2203.03821

摘要

We propose to explore a new problem called audio-visual segmentation (AVS), in which the goal is to output a pixel-level map of the object(s) that produce sound at the time of the image frame.
To facilitate this research, we construct the first audio-visual segmentation benchmark (AVSBench), providing pixel-wise annotations for the sound- ing objects in audible videos. Two settings are studied with this bench- mark:
1) semi-su