Can You Hear, Localize, and Segment Continually? An Exemplar-Free Continual Learning Benchmark for Audio-Visual Segmentation

• URL: http://arxiv.org/abs/2603.08967v1
• Date: Mon, 09 Mar 2026 21:58:14 GMT
• Title: Can You Hear, Localize, and Segment Continually? An Exemplar-Free Continual Learning Benchmark for Audio-Visual Segmentation
• Abstract summary: Real-world environments are inherently dynamic, causing audio and visual distributions to evolve over time.<n>We introduce the first continual learning benchmark for Audio-Visual, comprising four learning protocols across single-source and multi-source AVS datasets.<n>We mitigate catastrophic forgetting by introducing Low-Rank Anchoring (LRA), which stabilizes adapted weights based on loss sensitivity.
• Score: 9.480328219158254
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Audio-Visual Segmentation (AVS) aims to produce pixel-level masks of sound producing objects in videos, by jointly learning from audio and visual signals. However, real-world environments are inherently dynamic, causing audio and visual distributions to evolve over time, which challenge existing AVS systems that assume static training settings. To address this gap, we introduce the first exemplar-free continual learning benchmark for Audio-Visual Segmentation, comprising four learning protocols across single-source and multi-source AVS datasets. We further propose a strong baseline, ATLAS, which uses audio-guided pre-fusion conditioning to modulate visual feature channels via projected audio context before cross-modal attention. Finally, we mitigate catastrophic forgetting by introducing Low-Rank Anchoring (LRA), which stabilizes adapted weights based on loss sensitivity. Extensive experiments demonstrate competitive performance across diverse continual scenarios, establishing a foundation for lifelong audio-visual perception. Code is available at${}^{*}$\footnote{Paper under review} - \hyperlink{https://gitlab.com/viper-purdue/atlas}{https://gitlab.com/viper-purdue/atlas} \keywords{Continual Learning \and Audio-Visual Segmentation \and Multi-Modal Learning}

Related papers

• AudioMosaic: Contrastive Masked Audio Representation Learning [53.52371029884106]
  We introduce textbfAudioMosaic, a contrastive learning-based audio encoder for general audio understanding.<n>AudioMosaic constructs positive pairs by applying structured time-frequency masking to spectrogram patches.<n>Experiments show that AudioMosaic achieves state-of-the-art performance on several standard audio benchmarks.
  arXiv  Detail & Related papers  (2026-05-14T00:56:51Z)
• Learning What To Hear: Boosting Sound-Source Association For Robust Audiovisual Instance Segmentation [37.91678426119673]
  Existing methods suffer from visual bias stemming from two fundamental issues: uniform additive fusion prevents queries from specializing to different sound sources, and visual-only training objectives allow queries to converge to arbitrary salient objects.<n>We propose Audio-Centric Query Generation using cross-attention, enabling each query to selectively attend to distinct sound sources and carry sound-specific priors into visual decoding.
  arXiv  Detail & Related papers  (2025-09-26T02:31:17Z)
• Revisiting Audio-Visual Segmentation with Vision-Centric Transformer [60.83798235788669]
  Audio-Visual (AVS) aims to segment sound-producing objects in video frames based on the associated audio signal.<n>We propose a new Vision-Centric Transformer framework that leverages vision-derived queries to iteratively fetch corresponding audio and visual information.<n>Our framework achieves new state-of-the-art performances on three subsets of the AVSBench dataset.
  arXiv  Detail & Related papers  (2025-06-30T08:40:36Z)
• Audio Visual Segmentation Through Text Embeddings [17.285669984798975]
  Research on Audio-Visual (AVS) suffers from data scarcity due to the high cost of fine-grained manual annotations.<n>Recent works attempt to overcome the challenge of limited data by leveraging the vision foundation model, Segment Anything Model (SAM)<n>We propose textbfAV2T-SAM, a novel framework that bridges audio features with the text embedding space of pre-trained text-prompted SAM.
  arXiv  Detail & Related papers  (2025-02-22T21:15:44Z)
• Separating the "Chirp" from the "Chat": Self-supervised Visual Grounding of Sound and Language [77.33458847943528]
  We present DenseAV, a novel dual encoder grounding architecture that learns high-resolution, semantically meaningful, and audio-visually aligned features solely through watching videos. We show that DenseAV can discover the meaning'' of words and the location'' of sounds without explicit localization supervision.
  arXiv  Detail & Related papers  (2024-06-09T03:38:21Z)
• QDFormer: Towards Robust Audiovisual Segmentation in Complex Environments with Quantization-based Semantic Decomposition [47.103732403296654]
  Multi-source semantic space can be represented as the Cartesian product of single-source sub-spaces. We introduce a global-to-local quantization mechanism, which distills knowledge from stable global (clip-level) features into local (frame-level) ones. Experiments demonstrate that our semantically decomposed audio representation significantly improves AVS performance.
  arXiv  Detail & Related papers  (2023-09-29T20:48:44Z)
• Leveraging Foundation models for Unsupervised Audio-Visual Segmentation [49.94366155560371]
  Audio-Visual (AVS) aims to precisely outline audible objects in a visual scene at the pixel level. Existing AVS methods require fine-grained annotations of audio-mask pairs in supervised learning fashion. We introduce unsupervised audio-visual segmentation with no need for task-specific data annotations and model training.
  arXiv  Detail & Related papers  (2023-09-13T05:05:47Z)
• Class-Incremental Grouping Network for Continual Audio-Visual Learning [42.284785756540806]
  We propose a class-incremental grouping network (CIGN) that can learn category-wise semantic features to achieve continual audio-visual learning. We conduct extensive experiments on VGGSound-Instruments, VGGSound-100, and VGG-Sound Sources benchmarks. Our experimental results demonstrate that the CIGN achieves state-of-the-art audio-visual class-incremental learning performance.
  arXiv  Detail & Related papers  (2023-09-11T07:36:16Z)
• Text-to-feature diffusion for audio-visual few-shot learning [59.45164042078649]
  Few-shot learning from video data is a challenging and underexplored, yet much cheaper, setup. We introduce a unified audio-visual few-shot video classification benchmark on three datasets. We show that AV-DIFF obtains state-of-the-art performance on our proposed benchmark for audio-visual few-shot learning.
  arXiv  Detail & Related papers  (2023-09-07T17:30:36Z)
• AVSegFormer: Audio-Visual Segmentation with Transformer [42.24135756439358]
  A new audio-visual segmentation (AVS) task has been introduced, aiming to locate and segment the sounding objects in a given video. This task demands audio-driven pixel-level scene understanding for the first time, posing significant challenges. We propose AVSegFormer, a novel framework for AVS tasks that leverages the transformer architecture.
  arXiv  Detail & Related papers  (2023-07-03T16:37:10Z)
• SLICER: Learning universal audio representations using low-resource self-supervised pre-training [53.06337011259031]
  We present a new Self-Supervised Learning approach to pre-train encoders on unlabeled audio data. Our primary aim is to learn audio representations that can generalize across a large variety of speech and non-speech tasks.
  arXiv  Detail & Related papers  (2022-11-02T23:45:33Z)
• Audio-visual Generalised Zero-shot Learning with Cross-modal Attention and Language [38.02396786726476]
  We propose to learn multi-modal representations from audio-visual data using cross-modal attention. In our generalised audio-visual zero-shot learning setting, we include all the training classes in the test-time search space. Due to the lack of a unified benchmark in this domain, we introduce a (generalised) zero-shot learning benchmark on three audio-visual datasets.
  arXiv  Detail & Related papers  (2022-03-07T18:52:13Z)
• Learning Representations from Audio-Visual Spatial Alignment [76.29670751012198]
  We introduce a novel self-supervised pretext task for learning representations from audio-visual content. The advantages of the proposed pretext task are demonstrated on a variety of audio and visual downstream tasks.
  arXiv  Detail & Related papers  (2020-11-03T16:20:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.