Research map / Video / Video-language understanding
Action-and-object Aware Alignment for Partially Relevant Video Retrieval
Chuanshen Chen, Kai Zhou, Zhiquan Wen, Zeng You, Yirui Li, Tianhang Xiang, Mingkui Tan
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 |
|---|---|---|---|
| fuse action and object branches rather than using one (ablation) | A3PRVR-Base · Charades-STA | SumR: 64.7 -> 73.1 | Improved |
| add Q-DTA module (ablation) | A3PRVR-Base · Charades-STA | SumR: 73.1 -> 78.2 | Improved |
| add action- and object-aware contrastive losses LA, LO (ablation) | A3PRVR-Base · Charades-STA | SumR: 73.1 -> 76.3 | Improved |
| combine Q-DTA and LA, LO (ablation) | A3PRVR-Base · Charades-STA | SumR: 73.1 -> 82.3 | Improved |
| use vanilla cross-attention instead of Q-DTA (ablation) | A3PRVR-Base · Charades-STA | SumR: 73.1 -> 74.9 | Improved |
| use query-specific deformed points instead of shared deformed points (ablation) | Q-DTA · Charades-STA | SumR: 74.8 -> 78.2 | Improved |
| enhance object features with Q-DTA (O'') for matching (ablation) | A3PRVR-Base · Charades-STA | SumR: 73.1 -> 69.9 | Hurt |
| apply LA to object branch and LO to action branch (mismatch) (ablation) | A3PRVR-Base · Charades-STA | SumR: 76.3 -> 74.4 | No clear effect |
| A3PRVR vs AMDNet on ActivityNet-Caption | A3PRVR · ActivityNet-Caption | SumR: 172.8 -> 173.2 | No clear effect |
Closest papers by meaning
- Mitigating Semantic Collapse in Partially Relevant Video Retrieval (NeurIPS 2025 · similarity 0.89)
- Bridging the Semantic Granularity Gap Between Text and Frame Representations for Partially Relevant Video Retrieval (AAAI 2025 · similarity 0.89)
- MUVR: A Multi-Modal Untrimmed Video Retrieval Benchmark with Multi-Level Visual Correspondence (NeurIPS 2025 · similarity 0.87)
- MomentSeeker: A Task-Oriented Benchmark For Long-Video Moment Retrieval (NeurIPS 2025 · similarity 0.86)
- From Play to Replay: Composed Video Retrieval for Temporally Fine-Grained Videos (NeurIPS 2025 · similarity 0.86)
- Towards Balanced Alignment: Modal-Enhanced Semantic Modeling for Video Moment Retrieval (AAAI 2024 · similarity 0.86)
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.
