Research map / Video / Video-language understanding
Hierarchical Visual-Textual Graph for Temporal Activity Localization via Language
Shaoxiang Chen, Yu-Gang 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 object-sentence subgraph (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 40.38 -> 44.97 | Improved |
| add object-sentence subgraph (ablation) | HVTG · Charades-STA | Recall@IoU=0.7: 19.49 -> 21.77 | Improved |
| add object-object subgraph (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 40.38 -> 42.31 | Improved |
| add sentence-guided node aggregation (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 40.38 -> 42.93 | Improved |
| combine all three HVTG components (OSS, OOS, SGNA) (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 40.38 -> 47.27 | Improved |
| combine all three HVTG components (OSS, OOS, SGNA) (ablation) | HVTG · Charades-STA | Recall@IoU=0.3: 55.56 -> 61.37 | Improved |
| add cross-channel communication graph (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 43.23 -> 44.65 | Improved |
| add cross-channel communication graph on top of alignment loss (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 45.51 -> 47.27 | Improved |
| add alignment loss (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 43.23 -> 45.51 | Improved |
| add alignment loss on top of C3G (ablation) | HVTG · Charades-STA | Recall@IoU=0.5: 44.65 -> 47.27 | Improved |
| add alignment loss (ablation) | HVTG · Charades-STA | Recall@IoU=0.7: 20.77 -> 22.31 | Improved |
| add cross-channel communication graph (ablation) | HVTG · Charades-STA | Recall@IoU=0.7: 20.77 -> 22.42 | Improved |
Closest papers by meaning
- TALL: Temporal Activity Localization via Language Query (ICCV 2017 · similarity 0.91)
- Semantic Proposal for Activity Localization in Videos via Sentence Query (AAAI 2019 · similarity 0.88)
- Cross-Sentence Temporal and Semantic Relations in Video Activity Localisation (ICCV 2021 · similarity 0.88)
- Multi-Stage Aggregated Transformer Network for Temporal Language Localization in Videos (CVPR 2021 · similarity 0.87)
- ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos (CVPR 2025 · similarity 0.86)
- VicTR: Video-conditioned Text Representations for Activity Recognition (CVPR 2024 · similarity 0.86)
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.
