Research map / Video / Video-language understanding
Temporally Grounding Language Queries in Videos by Contextual Boundary-Aware Prediction
Jingwen Wang, Lin Ma, Wenhao Jiang
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 boundary prediction and modulate anchor scores with boundary scores (ablation) | CBP baseline · TACoS | R@1, IoU=0.7: 11.88 -> 16.02 | Improved |
| add contextual integration to the boundary model (ablation) | CBP + Boundary · TACoS | R@1, IoU=0.7: 16.02 -> 19.10 | Improved |
| replace self-attention context with concatenation-based context (ablation) | CBP full model · TACoS | mIoU: 21.59 -> 19.98 | Hurt |
| replace self-attention context with global context (ablation) | CBP full model · TACoS | mIoU: 21.59 -> 19.87 | Hurt |
| replace boundary prediction with offset regression (ablation) | CBP full model · TACoS | mIoU: 21.59 -> 20.79 | Hurt |
| use CBP rather than the strongest reported baseline | TACoS grounding · TACoS | R@1, IoU=0.7: 11.88 -> 19.10 | Improved |
| use CBP rather than the strongest reported baseline | Charades-STA grounding · Charades-STA | R@1, IoU=0.7: 15.80 -> 18.87 | Improved |
| use CBP rather than Xu et al. | Charades-STA grounding · Charades-STA | R@5, IoU=0.5: 79.40 -> 70.94 | Hurt |
| use CBP rather than the strongest reported baseline | ActivityNet Captions grounding · ActivityNet Captions | R@1, IoU=0.7: 13.93 -> 17.80 | Improved |
Closest papers by meaning
- Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Generation for Zero-Shot Natural Language Video Localization (AAAI 2025 · similarity 0.92)
- Context-Aware Biaffine Localizing Network for Temporal Sentence Grounding (CVPR 2021 · similarity 0.91)
- LLM-Powered Query Expansion for Enhancing Boundary Prediction in Language-Driven Action Localization (IJCV 2026 · similarity 0.91)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.88)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.88)
- ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos (CVPR 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.
