Audio-Visual World Models: Towards Multisensory Imagination in Sight and Sound

• URL: http://arxiv.org/abs/2512.00883v1
• Date: Sun, 30 Nov 2025 13:11:56 GMT
• Title: Audio-Visual World Models: Towards Multisensory Imagination in Sight and Sound
• Authors: Jiahua Wang, Shannan Yan, Leqi Zheng, Jialong Wu, Yaoxin Mao,
• Abstract summary: This work presents the first formal framework for Audio-Visual World Models (AVWM)<n>It formulates multimodal environment simulation as a partially observable decision process with audio-visual observations, fine-grained actions, and task rewards.<n>We propose an Audio-Visual Conditional Transformer with a novel modality expert architecture that balances visual and auditory learning.
• Score: 5.591620304505415
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: World models simulate environmental dynamics to enable agents to plan and reason about future states. While existing approaches have primarily focused on visual observations, real-world perception inherently involves multiple sensory modalities. Audio provides crucial spatial and temporal cues such as sound source localization and acoustic scene properties, yet its integration into world models remains largely unexplored. No prior work has formally defined what constitutes an audio-visual world model or how to jointly capture binaural spatial audio and visual dynamics under precise action control with task reward prediction. This work presents the first formal framework for Audio-Visual World Models (AVWM), formulating multimodal environment simulation as a partially observable Markov decision process with synchronized audio-visual observations, fine-grained actions, and task rewards. To address the lack of suitable training data, we construct AVW-4k, a dataset comprising 30 hours of binaural audio-visual trajectories with action annotations and reward signals across 76 indoor environments. We propose AV-CDiT, an Audio-Visual Conditional Diffusion Transformer with a novel modality expert architecture that balances visual and auditory learning, optimized through a three-stage training strategy for effective multimodal integration. Extensive experiments demonstrate that AV-CDiT achieves high-fidelity multimodal prediction across visual and auditory modalities with reward. Furthermore, we validate its practical utility in continuous audio-visual navigation tasks, where AVWM significantly enhances the agent's performance.

Related papers

• Audio-Guided Dynamic Modality Fusion with Stereo-Aware Attention for Audio-Visual Navigation [41.85539404067887]
  In audio-visual navigation (AVN) tasks, an embodied agent must autonomously localize a sound source in complex 3D environments.<n>Existing methods often rely on static modality fusion strategies and neglect the spatial cues embedded in stereo audio.<n>We propose an end-to-end reinforcement learning-based AVN framework with two key innovations.
  arXiv  Detail & Related papers  (2025-09-21T05:11:09Z)
• Spatial and Semantic Embedding Integration for Stereo Sound Event Localization and Detection in Regular Videos [3.2472293599354596]
  This report presents our systems submitted to the audio-only and audio-visual tracks of the DCASE2025 Task 3 Challenge: Stereo Sound Event localization and Detection in Regular Video Content.<n>SELD is a complex task that combines temporal event classification with spatial localization, requiring reasoning across spatial, temporal, and semantic dimensions.<n>To address this, we enhance standard SELD architectures with semantic information by integrating pre-trained, contrastive language-aligned models: CLAP for audio and OWL-ViT for visual inputs.
  arXiv  Detail & Related papers  (2025-07-07T10:08:57Z)
• Differentiable Room Acoustic Rendering with Multi-View Vision Priors [12.30408352143278]
  We introduce Audio-Visual Differentiable Room Acoustic Rendering (AV-DAR), a framework that leverages visual cues extracted from multi-view images and acoustic beam tracing for physics-based room acoustic rendering.<n>Experiments across six real-world environments from two datasets demonstrate that our multimodal, physics-based approach is efficient, interpretable, and accurate.
  arXiv  Detail & Related papers  (2025-04-30T17:55:29Z)
• AVadCLIP: Audio-Visual Collaboration for Robust Video Anomaly Detection [57.649223695021114]
  We present a novel weakly supervised framework that leverages audio-visual collaboration for robust video anomaly detection.<n>Our framework demonstrates superior performance across multiple benchmarks, with audio integration significantly boosting anomaly detection accuracy.
  arXiv  Detail & Related papers  (2025-04-06T13:59:16Z)
• Joint Multimodal Transformer for Emotion Recognition in the Wild [49.735299182004404]
  Multimodal emotion recognition (MMER) systems typically outperform unimodal systems. This paper proposes an MMER method that relies on a joint multimodal transformer (JMT) for fusion with key-based cross-attention.
  arXiv  Detail & Related papers  (2024-03-15T17:23:38Z)
• Unveiling the Power of Audio-Visual Early Fusion Transformers with Dense Interactions through Masked Modeling [24.346868432774453]
  Humans possess a remarkable ability to integrate auditory and visual information, enabling a deeper understanding of the surrounding environment. This early fusion of audio and visual cues, demonstrated through cognitive psychology and neuroscience research, offers promising potential for developing multimodal perception models. We address training early fusion architectures by leveraging the masked reconstruction framework, previously successful in unimodal settings, to train audio-visual encoders with early fusion. We propose an attention-based fusion module that captures interactions between local audio and visual representations, enhancing the model's ability to capture fine-grained interactions.
  arXiv  Detail & Related papers  (2023-12-02T03:38:49Z)
• AV-SUPERB: A Multi-Task Evaluation Benchmark for Audio-Visual Representation Models [92.92233932921741]
  We propose the AV-SUPERB benchmark that enables general-purpose evaluation of unimodal audio/visual and bimodal fusion representations. We evaluate 5 recent self-supervised models and show that none of these models generalize to all tasks. We show that representations may be improved with intermediate-task fine-tuning and audio event classification with AudioSet serves as a strong intermediate task.
  arXiv  Detail & Related papers  (2023-09-19T17:35:16Z)
• Improving Audio-Visual Segmentation with Bidirectional Generation [40.78395709407226]
  We introduce a bidirectional generation framework for audio-visual segmentation. This framework establishes robust correlations between an object's visual characteristics and its associated sound. We also introduce an implicit volumetric motion estimation module to handle temporal dynamics.
  arXiv  Detail & Related papers  (2023-08-16T11:20:23Z)
• AV-NeRF: Learning Neural Fields for Real-World Audio-Visual Scene Synthesis [61.07542274267568]
  We study a new task -- real-world audio-visual scene synthesis -- and a first-of-its-kind NeRF-based approach for multimodal learning. We propose an acoustic-aware audio generation module that integrates prior knowledge of audio propagation into NeRF. We present a coordinate transformation module that expresses a view direction relative to the sound source, enabling the model to learn sound source-centric acoustic fields.
  arXiv  Detail & Related papers  (2023-02-04T04:17:19Z)
• Joint Learning of Visual-Audio Saliency Prediction and Sound Source Localization on Multi-face Videos [101.83513408195692]
  We propose a multitask learning method for visual-audio saliency prediction and sound source localization on multi-face video. The proposed method outperforms 12 state-of-the-art saliency prediction methods, and achieves competitive results in sound source localization.
  arXiv  Detail & Related papers  (2021-11-05T14:35:08Z)
• Visually Informed Binaural Audio Generation without Binaural Audios [130.80178993441413]
  We propose PseudoBinaural, an effective pipeline that is free of recordings. We leverage spherical harmonic decomposition and head-related impulse response (HRIR) to identify the relationship between spatial locations and received audios. Our-recording-free pipeline shows great stability in cross-dataset evaluation and achieves comparable performance under subjective preference.
  arXiv  Detail & Related papers  (2021-04-13T13:07:33Z)
• Curriculum Audiovisual Learning [113.20920928789867]
  We present a flexible audiovisual model that introduces a soft-clustering module as the audio and visual content detector. To ease the difficulty of audiovisual learning, we propose a novel learning strategy that trains the model from simple to complex scene. We show that our localization model significantly outperforms existing methods, based on which we show comparable performance in sound separation without referring external visual supervision.
  arXiv  Detail & Related papers  (2020-01-26T07:08:47Z)

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.