CPM: Class-conditional Prompting Machine for Audio-visual Segmentation

TL;DR

This paper introduces CPM, a novel class-conditional prompting approach that enhances transformer-based audio-visual segmentation by improving cross-modal interaction and bipartite matching, achieving state-of-the-art results.

Contribution

The paper proposes CPM, a new framework that addresses training issues in AVS by combining class-agnostic and class-conditional queries and introducing new learning objectives.

Findings

Achieves state-of-the-art segmentation accuracy on AVS benchmarks.

Improves bipartite matching with a combined query learning strategy.

Enhances cross-modal attention effectiveness with new learning objectives.

Abstract

Audio-visual segmentation (AVS) is an emerging task that aims to accurately segment sounding objects based on audio-visual cues. The success of AVS learning systems depends on the effectiveness of cross-modal interaction. Such a requirement can be naturally fulfilled by leveraging transformer-based segmentation architecture due to its inherent ability to capture long-range dependencies and flexibility in handling different modalities. However, the inherent training issues of transformer-based methods, such as the low efficacy of cross-attention and unstable bipartite matching, can be amplified in AVS, particularly when the learned audio query does not provide a clear semantic clue. In this paper, we address these two issues with the new Class-conditional Prompting Machine (CPM). CPM improves the bipartite matching with a learning strategy combining class-agnostic queries with class-conditional queries. The efficacy of cross-modal attention is upgraded with new learning objectives for the audio, visual and joint modalities. We conduct experiments on AVS benchmarks, demonstrating that our method achieves state-of-the-art (SOTA) segmentation accuracy.