arrow
返回

Audio-Visual Segmentation with Semantics

delete2024-10-15
delete0
PRE
AI
J
Jinxing Zhou
X
Xuyang Shen
J
Jianyuan Wang
章
章嘉懿 (Jiayi Zhang)
W
Weixuan Sun
J
Jing Zhang
S
Stan Birchfield
D
Dan Guo
L
Lingpeng Kong
王
王萌 (Meng Wang) *
Y
Yiran Zhong *
DOI:10.1007/s11263-024-02261-xdelete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
We propose 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, i.e., AVSBench, providing pixel-wise annotations for sounding objects in audible videos. It contains three subsets: AVSBench-object (Single-source subset, Multi-sources subset) and AVSBench-semantic (Semantic-labels subset). Accordingly, three settings are studied: 1) semi-supervised audio-visual segmentation with a single sound source; 2) fully-supervised audio-visual segmentation with multiple sound sources, and 3) fully-supervised audio-visual semantic segmentation. The first two settings need to generate binary masks of sounding objects indicating pixels corresponding to the audio, while the third setting further requires to generate semantic maps indicating the object category. To deal with these problems, we propose a new baseline method that uses a temporal pixel-wise audio-visual interaction module to inject audio semantics as guidance for the visual segmentation process. We also design a regularization loss to encourage audio-visual mapping during training. Quantitative and qualitative experiments on the AVSBench dataset compare our approach to several existing methods for related tasks, demonstrating that the proposed method is promising for building a bridge between the audio and pixel-wise visual semantics. Code can be found at https://github.com/OpenNLPLab/AVSBench.
Keyword:
Audio-visual segmentation
Multi-modal segmentation
Audio-visual learning
AVSBench
Semantic segmentation
Video segmentation

期刊

International Journal of Computer Vision 封面图
International Journal of Computer Vision
IF:
9.3
论文数:
3.9K
被引数:
2.8W

机构

H
hefei university of technology
学者数:
2.5W
论文数: 1.7W
被引数: 35
U
University of Hong Kong
学者数:
4.1W
论文数: 3.9W
被引数: 10.1W
B
Beihang University
学者数:
5.2W
论文数: 4.1W
被引数: 37
A
Australian National University
学者数:
2.1W
论文数: 2.3W
被引数: 3.9W
N
nvidia corporation
学者数:
767
论文数: 439
被引数: 1
S
Shanghai Artificial Intelligence Laboratory
学者数:
475
论文数: 261
被引数: 765
U
university of oxford
学者数:
9.8W
论文数: 8.6W
被引数: 137
学者 查看更多机构
引用论文

引用论文

err1999-01-01
err0
PREAI
errD. Weygand; Y. Bréchet; J. Lépinoux
err分享
err收藏
Immuno-PET Imaging of HER3 in a Model in which HER3 Signaling Plays a Critical Role
err2015-11-16
err0
errOAAI
errQinghua Yuan; Takako Furukawa; Takahiro Tashiro; Kouki Okita; Zhao-Hui Jin; Winn Aung; Aya Sugyo; Kotaro Nagatsu; Hiroko Endo; Atsushi B. Tsuji; Ming-Rong Zhang; Takashi Masuko; Masahiro Inoue; Yasuhisa Fujibayashi; Tsuneo Saga
err分享
err收藏
Periphyton responses to nutrient and atrazine mixtures introduced through agricultural runoff
err2012-11-18
err0
errOAAI
errJustin N. Murdock; F. Douglas Shields; Richard E. Lizotte
err分享
err收藏
学者 查看更多内容