Research map / Vision-language and multimodal / Cross-modal learning
Label-anticipated Event Disentanglement for Audio-Visual Video Parsing
Jinxing Zhou, Dan Guo, Yuxin Mao, Yiran Zhong, Xiaojun Chang, Meng Wang
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 |
|---|---|---|---|
| replace MMIL decoder with LEAP label-projection decoder | MM-Pyr · LLP test, event-level Type@AV | F1: 57.1 -> 59.5 | Improved |
| replace MMIL decoder with LEAP label-projection decoder | HAN · LLP test, event-level Type@AV | F1: 56.4 -> 57.7 | Improved |
| add audio-visual semantic similarity loss L_avss (ablation) | LEAP with MM-Pyr · LLP test, event-level Event@AV | F1: 56.6 -> 57.4 | Improved |
| increase LEAP blocks from 1 to 2 (ablation) | LEAP with MM-Pyr · LLP test | average of ten metrics: 61.0 -> 61.3 | Improved |
| increase LEAP blocks from 2 to 4 (ablation) | LEAP with MM-Pyr · LLP test | average of ten metrics: 61.3 -> 61.4 | No clear effect |
| swap Glove label embeddings for Bert or CLIP (ablation) | LEAP with MM-Pyr · LLP test | average of ten metrics: 61.3 -> 60.9 (Bert), 61.1 (CLIP) | No clear effect |
| use last block output rather than first or average (ablation) | LEAP with MM-Pyr · LLP test | average of ten metrics: 60.7 (first), 60.8 (average) -> 61.3 (last) | Improved |
| set lambda to 1.0 instead of 0.5 or 2.0 (ablation) | LEAP with MM-Pyr · LLP test | average of ten metrics: 61.8 (0.5), 61.5 (2.0) -> 62.1 | Improved |
| use LEAP in place of the default decoder on overlapping events | MM-Pyr · LLP overlapping subset, event-level | average F1: 48.1 -> 49.8 | Improved |
| replace default decoder with LEAP on audio-visual event localization | CMBS · AVE dataset | accuracy: 74.5 -> 77.9 | Improved |
Closest papers by meaning
- Weakly-Supervised Audio-Visual Video Parsing with Prototype-based Pseudo-Labeling (CVPR 2024 · similarity 0.92)
- Multimodal Class-aware Semantic Enhancement Network for Audio-Visual Video Parsing (AAAI 2025 · similarity 0.92)
- Audio-Visual Semantic Graph Network for Audio-Visual Event Localization (CVPR 2025 · similarity 0.90)
- UWAV: Uncertainty-weighted Weakly-supervised Audio-Visual Video Parsing (CVPR 2025 · similarity 0.90)
- PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling (NeurIPS 2025 · similarity 0.90)
- Adapting to the Unknown: Training-Free Audio-Visual Event Perception with Dynamic Thresholds (CVPR 2025 · similarity 0.90)
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.
