Preprint
Aug 2026
Acoustically Grounded Cost Learning for Open-Vocabulary Audio-Visual Semantic Segmentation
A novel Acoustically Grounded Cost Learning (AGCL) framework is proposed to transform the static, audio-agnostic visual-text priors into dynamic, audio-grounded cost representations and significantly outperforms previous state-of-the-art approaches, particularly on unseen categories.
Tianrui Hui, Shao-Fei Huang, Qi-Song Han et al.
· 0 citations