Research map / Video / Video-language understanding
Text-Based Temporal Localization of Novel Events
Sudipta Paul, Niluthpol Chowdhury Mithun, Amit K. Roy-Chowdhury
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 semantically relevant seen moment as support instead of irrelevant one (ablation) | TLRR · Charades-STA Unseen | R@1 IoU@0.5: 20.30 -> 33.15 | Improved |
| use semantically relevant seen moment as support instead of random one (ablation) | TLRR · Charades-STA Unseen | R@1 IoU@0.5: 28.71 -> 33.15 | Improved |
| use relevant instead of irrelevant support (ablation) | TLRR · Charades-STA Unseen | mIoU: 22.48 -> 31.29 | Improved |
| use relevant instead of random support (ablation) | TLRR · Charades-STA Unseen | R@5 IoU@0.7: 40.24 -> 42.40 | Improved |
| add negative relational loss Lneg to Lintra (ablation) | TLRR · Charades-STA Unseen | support relevance sensitivity: within 1% -> 5% -15% difference | Improved |
| replace 2D-TAN with TLRR | 2D-TAN · Charades-STA Unseen | R@1 IoU@0.7: 13.33 -> 16.22 | Improved |
| replace 2D-TAN with TLRR | 2D-TAN · ActivityNet Captions Unseen | R@5 IoU@0.7: 24.01 -> 36.66 | Improved |
| replace SCDM with TLRR | SCDM · ActivityNet Captions Unseen | R@5 IoU@0.5: 46.38 -> 53.31 | Improved |
| use TLRR instead of SCDM for seen-vs-unseen gap | SCDM · Charades-STA Unseen | avg gap: 19.80 -> 3.37 | Improved |
| use TLRR instead of 2D-TAN for seen-vs-unseen gap | 2D-TAN · ActivityNet Captions Unseen | avg gap: 16.18 -> 7.13 | Improved |
| use TLRR instead of 2D-TAN on seen queries | 2D-TAN · Charades-STA Unseen (seen queries) | R@1 IoU@0.5: 37.95 -> 34.83 | Mixed |
| use TLRR instead of SCDM on seen queries | SCDM · ActivityNet Captions Unseen (seen queries) | R@1 IoU@0.5: 34.66 -> 27.46 | Hurt |
Closest papers by meaning
- Transferable Video Moment Localization by Moment-Guided Query Prompting (AAAI 2024 · similarity 0.92)
- Localizing Moments in Video With Natural Language (ICCV 2017 · similarity 0.90)
- To Find Where You Talk: Temporal Sentence Localization in Video with Attention Based Location Regression (AAAI 2019 · similarity 0.89)
- Multi-Scale 2D Temporal Adjacency Networks for Moment Localization With Natural Language (TPAMI 2021 · similarity 0.89)
- The Devil is in the Spurious Correlations: Boosting Moment Retrieval with Dynamic Learning (ICCV 2025 · similarity 0.88)
- Localizing Events in Videos with Multimodal Queries (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.
