Research map / Video / Video-language understanding
End-to-End Dense Video Captioning With Parallel Decoding
Teng Wang, Ruimao Zhang, Zhichao Lu, Feng Zheng, Ran Cheng, Ping Luo
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 deformable instead of vanilla transformer (ablation) | PDVC_light · ActivityNet Captions val (ablation) | SODA_c: 3.06 -> 5.17 | Improved |
| use deformable instead of vanilla transformer (ablation) | PDVC_light · ActivityNet Captions val (ablation) | METEOR: 6.10 -> 7.11 | Improved |
| use LSTM-DSA captioner instead of plain LSTM (ablation) | PDVC · ActivityNet Captions val (ablation) | METEOR: 7.11 -> 7.50 | Improved |
| use LSTM-DSA captioner instead of plain LSTM (ablation) | PDVC · ActivityNet Captions val (ablation) | SODA_c: 5.17 -> 5.26 | Improved |
| use self-attention captioning head instead of LSTM (ablation) | PDVC · ActivityNet Captions val (ablation) | SODA_c: 5.17 -> 3.40 | Hurt |
| add event counter (ablation) | PDVC_light · ActivityNet Captions val (ablation) | SODA_c: 4.30 -> 5.17 | Improved |
| add event counter (ablation) | PDVC_light · ActivityNet Captions val (ablation) | Precision: 44.88 -> 57.65 | Improved |
| increase event queries from 10 to 100 (ablation) | PDVC_light · ActivityNet Captions val (ablation) | SODA_c: 5.17 -> 4.59 | Hurt |
| increase event queries from 10 to 100 (ablation) | PDVC_light · ActivityNet Captions val (ablation) | Recall: 55.92 -> 51.88 | Hurt |
| length modulation gamma 2.0 instead of 0.0 (summing) (ablation) | PDVC_light · ActivityNet Captions val (ablation) | METEOR: 5.78 -> 7.11 | Improved |
| length modulation gamma 2.0 instead of 1.0 (averaging) (ablation) | PDVC_light · ActivityNet Captions val (ablation) | Precision: 53.31 -> 57.65 | Improved |
| length modulation gamma 2.0 instead of 1.0 (averaging) (ablation) | PDVC_light · ActivityNet Captions val (ablation) | METEOR: 7.35 -> 7.11 | Hurt |
Closest papers by meaning
- Implicit Location-Caption Alignment via Complementary Masking for Weakly-Supervised Dense Video Captioning (AAAI 2025 · similarity 0.91)
- Event-Equalized Dense Video Captioning (CVPR 2025 · similarity 0.91)
- Do You Remember? Dense Video Captioning with Cross-Modal Memory Retrieval (CVPR 2024 · similarity 0.90)
- Streaming Dense Video Captioning (CVPR 2024 · similarity 0.89)
- Dense Video Object Captioning from Disjoint Supervision (ICLR 2025 · similarity 0.89)
- DIBS: Enhancing Dense Video Captioning with Unlabeled Videos via Pseudo Boundary Enrichment and Online Refinement (CVPR 2024 · similarity 0.88)
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.
