Research map / Probabilistic and scientific modeling / Variational and generative sampling
Variational Bayesian Inference for Audio-Visual Tracking of Multiple Speakers
Yutong Ban, Xavier Alameda-Pineda, Laurent Girin B, Radu Horaud
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use VAVIT instead of audio-visual baselines requiring simultaneous modalities | audio-visual tracking · AVDIAR living-room (FFOV) | MOTA: 18.96 -> 96.03 | Improved |
| use VAVIT instead of the best baseline when the camera field of view is partial | VAVIT · AVDIAR living-room (PFOV) | MOTA: 66.39 -> 69.62 | Improved |
| use VAVIT instead of the best baseline when the camera field of view is partial | VAVIT · AVDIAR meeting-room (PFOV) | MOTA: 64.24 -> 65.27 | Improved |
| use VAVIT instead of the visual-only tracker with full field of view | VAVIT · AVDIAR living-room (FFOV) | MOTA: 96.32 -> 96.03 | No clear effect |
| use VAVIT instead of the visual-only tracker with full field of view | VAVIT · AVDIAR meeting-room (FFOV) | MOTA: 98.50 -> 98.16 | No clear effect |
| use VAVIT instead of the second-best baseline on AV16.3 | VAVIT · AV16.3 | MOTA: 82.9 -> 84.1 | Improved |
| add audio observations to visual tracking in the full field of view | VAVIT · AVDIAR (FFOV) | OSPA-T: 14.79 -> 17.05 | Hurt |
| use VAVIT audio-visual tracking when persons leave the camera field of view | VAVIT · AVDIAR living-room (PFOV) | MT: 45% -> 70% | Improved |
| use VAVIT variational inference instead of particle-filter baseline in running time | VAVIT · AVDIAR living-room | computation time (s): 20821 -> 3456 | Improved |
| use VAVIT instead of visual-only OBVT in running time | VAVIT · AVDIAR living-room | computation time (s): 542 -> 3456 | Hurt |
| use VAVIT for diarization instead of audio-only DiarTK | VAVIT · AVDIAR diarization | DER: 42.58 -> 28.73 | Improved |
| use VAVIT for diarization instead of the dedicated HMM audio-visual diarization method | VAVIT · AVDIAR diarization | DER: 18.88 -> 28.73 | Hurt |
Closest papers by meaning
- Robust Audio-Visual Segmentation via Audio-Guided Visual Convergent Alignment (CVPR 2025 · similarity 0.85)
- Seeing Speech and Sound: Distinguishing and Locating Audio Sources in Visual Scenes (CVPR 2025 · similarity 0.84)
- A Versatile Diffusion Transformer with Mixture of Noise Levels for Audiovisual Generation (NeurIPS 2024 · similarity 0.84)
- Adapting to the Unknown: Training-Free Audio-Visual Event Perception with Dynamic Thresholds (CVPR 2025 · similarity 0.84)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (AAAI 2025 · similarity 0.84)
- MMDisCo: Multi-Modal Discriminator-Guided Cooperative Diffusion for Joint Audio and Video Generation (ICLR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
