Research map / Video / Video-language understanding
Unified Static and Dynamic Network: Efficient Temporal Filtering for Video Grounding
Jingjing Hu, Dan Guo, Kun Li, Zhan Si, Xun Yang, Xiaojun Chang, Meng Wang
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 |
|---|---|---|---|
| train with multiple queries per video (M) instead of single query (S) | UniSDNet (GloVe) · ActivityNet Captions | R@1 IoU@0.5: 52.73 -> 57.67 | Improved |
| train with multiple queries per video instead of single query | UniSDNet (BERT) · TACoS | R@1 IoU@0.5: 36.24 -> 40.26 | Improved |
| train with multiple queries per video instead of single query | UniSDNet · ActivityNet Speech | R@1 IoU@0.5: 47.82 -> 56.29 | Improved |
| train with multiple queries per video instead of single query | UniSDNet · TACoS Speech (C3D) | R@1 IoU@0.5: 31.77 -> 37.77 | Improved |
| train with multiple queries per video instead of single query | UniSDNet · Charades-STA Speech (I3D) | R@1 IoU@0.5: 53.63 -> 53.82 | Improved |
| train with multiple queries per video instead of single query | UniSDNet · Charades-STA (C3D) | R@1 IoU@0.7: 27.31 -> 28.39 | Improved |
| use BERT instead of GloVe text features | UniSDNet-M · ActivityNet Captions | R@1 IoU@0.5: 57.67 -> 60.75 | Improved |
| use BERT instead of GloVe text features | UniSDNet-S · ActivityNet Captions | R@1 IoU@0.5: 52.73 -> 52.35 | No clear effect |
| use BERT instead of GloVe text features | UniSDNet-M · TACoS | R@1 IoU@0.5: 38.34 -> 40.26 | Improved |
| use UniSDNet-M instead of MMN with same 2D temporal map | SLVG · Charades-STA Speech (VGG) | R@1 IoU@0.5: 42.74 -> 46.37 | Improved |
| use UniSDNet-S instead of VSLNet reproduced | SLVG · ActivityNet Speech (C3D) | mIoU: 37.04 -> 47.31 | Improved |
| use UniSDNet-M instead of MSAT | NLVG · TACoS | R@1 IoU@0.5: 37.57 -> 40.26 | Improved |
Closest papers by meaning
- SE-DETR: Explicit Semantic Exploration for Generalizability and Distinguishability in Video Temporal Grounding (ECCV 2026 · similarity 0.90)
- Keeping the Evidence Chain: Semantic Evidence Allocation for Training-Free Token Pruning in Video Temporal Grounding (ECCV 2026 · similarity 0.89)
- HERO: Hierarchical Embedding-Refinement for Open-Vocabulary Temporal Sentence Grounding in Videos (CVPR 2026 · similarity 0.89)
- Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding (NeurIPS 2025 · similarity 0.89)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.89)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · 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.
