# LightAVSeg: Lightweight Audio-Visual Segmentation

Source: [LightAVSeg: Lightweight Audio-Visual Segmentation](<https://www.sota2.com/research/paper/6a02b403dbc836b87d8245ce>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-13T05:00:05.563Z

Authors: Qing Zhong, Guodong Ding, Lingqiao Liu, Zaiwen Feng, Lin Yuanbo Wu, Angela Yao

Published: 2026-05-09T08:47:54.000Z

## Abstract

Audio-Visual Segmentation (AVS) targets pixel level localization of sounding emitting objects in videos. However, existing models rely on dense cross-modal attention with quadratic computational cost, limiting their suitability for resource efficient deployment. Most efficiency oriented methods focus on backbone reduction and overlook the interaction module as the primary bottleneck. This paper proposes LightAVSeg, a lightweight framework that replaces heavy attention with a decoupled design for semantic filtering and spatial grounding, resulting in interaction costs that scale linearly with spatial resolution. Furthermore, we introduce an auxiliary alignment loss to enforce semantic consistency during training with zero inference overhead. Extensive experiments demonstrate that LightAVSeg achieves a new state-of-the-art among lightweight methods: with 20.5M parameters ~1/7 of AVSegFormer), it reaches 50.4 mIoU on the MS3 benchmark and enables efficient inference on a mobile processor.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Audio-Visual Segmentation on AVSBench MS3](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-avsbench-ms3.md>): Audio-Visual Segmentation; AVSBench MS3; MJ Score: 50.4.

- [Audio-Visual Semantic Segmentation on AVSS](<https://www.sota2.com/research/sota/audio-visual-semantic-segmentation-on-avss.md>): Audio-Visual Semantic Segmentation; AVSS; MJ Score: 30.6.

- [Audio-Visual Segmentation on AVSBench S4](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-avsbench-s4.md>): Audio-Visual Segmentation; AVSBench S4; MJ Score: 75.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.08805>)

[PDF](<https://arxiv.org/pdf/2605.08805.pdf>)