Research map / Video / Video-language understanding
Bidirectional Attentive Fusion With Context Gating for Dense Video Captioning
Jingwen Wang, Wenhao Jiang, Lin Ma, Wei Liu, Yong Xu
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 |
|---|---|---|---|
| combine forward and backward proposal predictions to use past and future context | SST proposals · ActivityNet Captions | F1@1000: 0.566 -> 0.602 | Improved |
| combine forward and backward proposal predictions to use past and future context | SST proposals · THUMOS-14 | Recall@10: 0.053 -> 0.063 | Improved |
| add bidirectional proposal predictions | SST proposals · THUMOS-14, tIoU=0.8 | Recall@1000: 0.696 -> 0.711 | Improved |
| use bidirectional rather than forward-only proposal context (ablation) | ActivityNet Captions captioning · ActivityNet Captions validation, top 1000 proposals | Meteor: 7.87 -> 8.35 | Improved |
| add event clip features to context vectors (ablation) | Bi-SST captioning · ActivityNet Captions validation, top 1000 proposals | Meteor: 8.35 -> 8.78 | Improved |
| add context vectors to event clip features (ablation) | Bi-SST captioning · ActivityNet Captions validation, top 1000 proposals | Meteor: 8.36 -> 8.78 | Improved |
| use temporal dynamic attention instead of mean pooling (ablation) | Bi-SST + E + H · ActivityNet Captions validation, top 1000 proposals | Meteor: 8.78 -> 9.00 | Improved |
| add context gating to balance event features and context (ablation) | Bi-SST + E + H + TDA · ActivityNet Captions validation, top 1000 proposals | Meteor: 9.00 -> 9.19 | Improved |
| add joint proposal-caption ranking at inference (ablation) | Bi-SST + E + H + TDA + CG · ActivityNet Captions validation, top 1000 proposals | Meteor: 9.19 -> 9.60 | Improved |
| use the complete proposed system rather than the prior state-of-the-art method | Dense video captioning · ActivityNet Captions test | Meteor: 4.82 -> 9.65 | Improved |
Closest papers by meaning
- Event-Equalized Dense Video Captioning (CVPR 2025 · similarity 0.90)
- Implicit Location-Caption Alignment via Complementary Masking for Weakly-Supervised Dense Video Captioning (AAAI 2025 · similarity 0.88)
- Do You Remember? Dense Video Captioning with Cross-Modal Memory Retrieval (CVPR 2024 · similarity 0.88)
- Dense Video Object Captioning from Disjoint Supervision (ICLR 2025 · similarity 0.88)
- Progress-Aware Video Frame Captioning (CVPR 2025 · similarity 0.87)
- DIBS: Enhancing Dense Video Captioning with Unlabeled Videos via Pseudo Boundary Enrichment and Online Refinement (CVPR 2024 · similarity 0.87)
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.
