Research map / Video / Video-language understanding
Dense-Captioning Events in Videos
Ranjay Krishna, Kenji Hata, Frederic Ren, Li Fei-Fei, Juan Carlos Niebles
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 |
|---|---|---|---|
| add past-event context (ablation) | captioning module · ActivityNet Captions (GT proposals) | CIDEr: 20.97 -> 22.94 | Improved |
| add future-event context to online model (ablation) | captioning module · ActivityNet Captions (GT proposals) | CIDEr: 22.94 -> 24.56 | Improved |
| add attention over neighbor events instead of mean pooling (ablation) | full model · ActivityNet Captions (GT proposals) | CIDEr: 24.24 -> 24.56 | Improved |
| add attention over neighbor events instead of mean pooling (ablation) | online model · ActivityNet Captions (GT proposals) | CIDEr: 22.19 -> 22.94 | Improved |
| add attention instead of mean pooling (ablation) | full model · ActivityNet Captions (learnt proposals) | CIDEr: 15.29 -> 17.29 | Improved |
| add full context (ablation) | captioning module · ActivityNet Captions (learnt proposals) | CIDEr: 12.34 -> 17.29 | Improved |
| add future context (ablation) | captioning module · ActivityNet Captions first sentence | CIDEr: 31.56 -> 29.92 | Hurt |
| add future context (ablation) | captioning module · ActivityNet Captions third sentence | CIDEr: 19.36 -> 20.01 | Improved |
| use multiple strides in proposal module instead of stride 1 (ablation) | proposal module · ActivityNet Captions localization | recall: plateaus around a recall of 0.3 with stride 1 | Improved |
| use multiple strides with few proposals (ablation) | proposal module · ActivityNet Captions localization | recall | Hurt |
| add context from past events (ablation) | retrieval model · ActivityNet Captions video retrieval | median rank: 78 -> 36 | Improved |
| add future context to online retrieval model (ablation) | retrieval model · ActivityNet Captions video retrieval | median rank: 36 -> 34 | No clear effect |
Closest papers by meaning
- Event-Equalized Dense Video Captioning (CVPR 2025 · similarity 0.91)
- Do You Remember? Dense Video Captioning with Cross-Modal Memory Retrieval (CVPR 2024 · similarity 0.89)
- Exploring Temporal Event Cues for Dense Video Captioning in Cyclic Co-Learning (AAAI 2025 · similarity 0.88)
- Dense Video Object Captioning from Disjoint Supervision (ICLR 2025 · similarity 0.87)
- Streaming Dense Video Captioning (CVPR 2024 · similarity 0.86)
- DiffDVC: Accurate Event Detection for Dense Video Captioning via Diffusion Models (AAAI 2025 · 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.
