Research map / Video / Video-language understanding
Temporal Calibrating and Distilling for Scene-Text Aware Text-Video Retrieval
Zhiqian Zhao, Liang Li, Lei Shen, Xichun Sheng, Yaoqi Sun, Fang Kang, Chenggang Yan
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 window-OCR captioner | CLIP4Clip · TextVR | T2V R@1: 25.4 -> 33.5 | Improved |
| add window-OCR captioner | T-MASS · TextVR | T2V R@1: 28.4 -> 45.0 | Improved |
| add window-OCR captioner | T-MASS · M4-ViteVQA | T2V R@1: 44.5 -> 65.6 | Improved |
| add OCR (with OCR vs without) | TCD · MSR-VTT | T2V R@1: 54.5 -> 59.7 | Improved |
| add co-attention (ablation) | TCD · MSR-VTT (OCR-free) | R@1: 49.8 -> 54.5 | Improved |
| add HSC loss Lhsc (ablation) | TCD · MSR-VTT (OCR-free) | R@1: 53.9 -> 54.5 | Improved |
| add CTCD (ablation) | TCD · MSR-VTT (OCR-free) | R@1: 49.3 -> 54.5 | Improved |
| add text condition (ablation) | TCD · MSR-VTT (OCR-free) | R@1: 38.6 -> 54.5 | Improved |
| add CTCD, co-attention and Lhsc together (ablation) | TCD · TextVR | T2V R@1: 53.3 -> 60.4 | Improved |
| use hybrid clue (C*_V + C*_O)/2 instead of cross-attention (ablation) | CTCD · TextVR | T2V R@1: 58.6 -> 60.4 | Improved |
| add OCR-domain clue C*_O to video clue (ablation) | CTCD · TextVR | T2V R@1: 59.4 -> 60.4 | Improved |
| replace contrastive-learning alignment with HSC (ablation) | TCD · TextVR | R@1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- UATVR — UATVR: Uncertainty-Adaptive Text-Video Retrieval (ICCV 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SViTT: Temporal Learning of Sparse Video-Text Transformers (CVPR 2023 · similarity 0.88)
- Towards Generalisable Video Moment Retrieval: Visual-Dynamic Injection to Image-Text Pre-Training (CVPR 2023 · similarity 0.88)
- Learning Text-to-Video Retrieval from Image Captioning (IJCV 2024 · similarity 0.88)
- Heterogeneous Prompt-Guided Entity Inferring and Distilling for Scene-Text Aware Cross-Modal Retrieval (AAAI 2025 · similarity 0.88)
- Logic-in-Frames: Dynamic Keyframe Search via Visual Semantic-Logical Verification for Long Video Understanding (NeurIPS 2025 · similarity 0.87)
- TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval (ICLR 2025 · similarity 0.87)
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.
