Research map / Video / Video-language understanding
Open Set Video HOI detection from Action-Centric Chain-of-Look Prompting
Nan Xi, Jingjing Meng, Junsong Yuan
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 VideoHOI Prompting (visual feature chain) (ablation) | ACoLP · VidHOI (closed set) | mAP: 19.45 -> 28.27 | Improved |
| add CaptionHOI Prompting (BLIP caption chain) (ablation) | ACoLP · VidHOI (closed set) | mAP: 22.06 -> 28.27 | Improved |
| add Action Prompting module (ablation) | ACoLP · VidHOI (closed set) | mAP: 21.94 -> 28.27 | Improved |
| add Dynamic GNN for temporal modeling (ablation) | ACoLP · VidHOI (closed set) | mAP: 20.67 -> 28.27 | Improved |
| add VideoHOI Prompting (ablation) | ACoLP · VidHOI open set (20% unseen) | mAP: 14.38 -> 19.23 | Improved |
| add CaptionHOI Prompting (ablation) | ACoLP · VidHOI open set (50% unseen) | mAP: 11.57 -> 12.78 | Improved |
| add VideoHOI Prompting (ablation) | ACoLP · CAD120 open set (50% unseen) | F1: 74.56 -> 76.39 | Improved |
| action-centric model versus TUTOR on temporal-related HOIs | ACoLP · VidHOI temporal-related HOIs | mAP: 21.28 -> 24.77 | Improved |
| ACoLP versus TUTOR on static spatial HOIs | ACoLP · VidHOI spatial-related HOIs | mAP: 32.21 -> 32.96 | Improved |
| ACoLP versus TUTOR on rare categories | ACoLP · VidHOI rare categories | mAP: 23.49 -> 26.63 | Improved |
Closest papers by meaning
- Open-Vocabulary HOI Detection with Interaction-aware Prompt and Concept Calibration (ICCV 2025 · similarity 0.91)
- Open-World Human-Object Interaction Detection via Multi-modal Prompts (CVPR 2024 · similarity 0.91)
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 2024 · similarity 0.89)
- Toward Open-Set Human Object Interaction Detection (AAAI 2024 · similarity 0.89)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.89)
- EZ-HOI: VLM Adaptation via Guided Prompt Learning for Zero-Shot HOI Detection (NeurIPS 2024 · 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.
