Research map / Video / Video-language understanding
Support-Set Based Cross-Supervision for Video Grounding
Xinpeng Ding, Nannan Wang, Shiwei Zhang, De Cheng, Xiaomeng Li, Ziyuan Huang, Mingqian Tang, Xinbo Gao
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 support-set cross-supervision (contrastive+caption) to 2D-TAN | 2D-TAN · Charades-STA (I3D) | Rank1@0.5 (%): 50.62 -> 56.97 | Improved |
| add support-set cross-supervision to LGI | LGI · Charades-STA (I3D) | Rank1@0.5 (%): 59.46 -> 60.75 | Improved |
| add support-set cross-supervision to 2D-TAN | 2D-TAN · TACoS | Rank1@0.5 (%): 25.32 -> 29.56 | Improved |
| add support-set cross-supervision to 2D-TAN | 2D-TAN · ActivityNet-Captions | Rank1@0.5 (%): 44.51 -> 46.67 | Improved |
| add support-set cross-supervision to LGI | LGI · ActivityNet-Captions | Rank1@0.5 (%): 41.51 -> 43.62 | Improved |
| use support-set supervision instead of GT-clip supervision (both losses) (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 55.40 -> 56.97 | Improved |
| add contrastive loss to caption-only GT-clip supervision (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 51.72 -> 54.77 | Improved |
| add caption loss to contrastive-only support-set supervision (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 56.19 -> 56.97 | Improved |
| add contrastive loss to caption-only support-set supervision (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 53.12 -> 56.97 | Improved |
| add contrastive loss to caption-only support-set supervision (ablation) | LGI · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 60.53 -> 60.75 | No clear effect |
| build support set from whole video (V-SS) instead of GT clips only (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 55.91 -> 56.97 | Improved |
| build support set from whole video (V-SS) instead of non-GT clips only (ablation) | 2D-TAN · Charades-STA (I3D, ablation) | Rank1@0.5 (%): 50.12 -> 56.97 | Improved |
Closest papers by meaning
- Comprehensive Visual Grounding for Video Description (AAAI 2024 · similarity 0.86)
- Local-Global Multi-Modal Distillation for Weakly-Supervised Temporal Video Grounding (AAAI 2024 · similarity 0.86)
- Exploiting Auxiliary Caption for Video Grounding (AAAI 2024 · similarity 0.86)
- SnAG: Scalable and Accurate Video Grounding (CVPR 2024 · similarity 0.86)
- Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Video Temporal Grounding (NeurIPS 2025 · similarity 0.86)
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.85)
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.
