Research map / 3D vision / Human pose estimation
Turbo Learning Framework for Human-Object Interactions Recognition and Human Pose Estimation
Wei Feng, Wentao Liu, Tong Li, Jing Peng, Chen Qian, Xiaolin Hu
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 pose estimation features as input to HOI recognition, rather than using joint multi-task training without pose features (ablation) | Turbo Learning Framework · V-COCO | AP_agent: +0.2% | Improved |
| Add pose estimation features as input to HOI recognition, rather than using joint multi-task training without pose features (ablation) | Turbo Learning Framework · V-COCO | AP_role: +0.8% | Improved |
| Add a pose estimation branch and pose features to HOI recognition (ablation) | Turbo Learning Framework · V-COCO | AP_agent: +0.5% | Improved |
| Add a pose estimation branch and pose features to HOI recognition (ablation) | Turbo Learning Framework · V-COCO | AP_role: +1.1% | Improved |
| Add HOI recognition features as guidance to pose estimation (ablation) | Turbo Learning Framework · V-COCO | AP_50^kp: 1% | Improved |
| Add the second turbo stage after stage 1 (ablation) | Turbo Learning Framework · V-COCO | AP_agent: +1.7% | Improved |
| Add the second turbo stage after stage 1 (ablation) | Turbo Learning Framework · V-COCO | AP_role: +0.2% | Improved |
| Add the second turbo stage after stage 1 (ablation) | Turbo Learning Framework · V-COCO | AP_50^kp: +0.4% | Improved |
| Add the third turbo stage after stage 2 (ablation) | Turbo Learning Framework · V-COCO | AP_agent: +0.9% | Improved |
| Add the third turbo stage after stage 2 (ablation) | Turbo Learning Framework · V-COCO | AP_role: +1.3% | Improved |
| Add the third turbo stage after stage 2 (ablation) | Turbo Learning Framework · V-COCO | AP_50^kp: +0.8% | Improved |
| Pre-train on data with HOI and pose annotations, then fine-tune without keypoint loss (ablation) | Proposed (Pre-train) · HICO-DET test, Full | mAP: 9.94 -> 10.9 | Improved |
Closest papers by meaning
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.89)
- Detecting Any Human-Object Interaction Relationship: Universal HOI Detector with Spatial Prompt Learning on Foundation Models (NeurIPS 2023 · similarity 0.87)
- Reformulating HOI Detection As Adaptive Set Prediction (CVPR 2021 · similarity 0.87)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.85)
- HOIMamba: Efficient Mamba-based Disentangled Progressive Learning for HOI Detection (AAAI 2025 · similarity 0.85)
- Dual-Prior Augmented Decoding Network for Long Tail Distribution in HOI Detection (AAAI 2024 · similarity 0.84)
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.
