Research map / Detection and segmentation / Open-set and few-shot detection
HOI Analysis: Integrating and Decomposing Human-Object Interaction
Yong-Lu Li, Xinpeng Liu, Xiaoqian Wu, Yizhuo Li, Cewu Lu
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 inter-pair transformation (instance exchange) (ablation) | IDN · HICO-DET Default Full | mAP: 22.63 -> 23.36 | Improved |
| add integration validity loss L_cls^u (ablation) | IDN · HICO-DET Default Full | mAP: 19.98 -> 23.36 | Improved |
| add decomposition validity loss L_cls^ho (ablation) | IDN · HICO-DET Default Full | mAP: 21.39 -> 23.36 | Improved |
| add interactiveness validity loss L_bin (ablation) | IDN · HICO-DET Default Full | mAP: 22.01 -> 23.36 | Improved |
| add AE verb classification pre-training loss (ablation) | IDN · HICO-DET Default Full | mAP: 19.60 -> 23.36 | Improved |
| add AE reconstruction loss (ablation) | IDN · HICO-DET Default Full | mAP: 21.07 -> 23.36 | Improved |
| combine integration, decomposition and AE scores instead of integration only (ablation) | IDN · HICO-DET Default Full | mAP: 21.26 -> 23.36 | Improved |
| use per-verb MLP pairs instead of single conditioned MLP with verb indicator (ablation) | IDN · HICO-DET Default Full | mAP: 20.86 -> 23.36 | Improved |
| use loop through integrated features instead of using f_u directly (ablation) | IDN · HICO-DET Default Full | mAP: 21.77 -> 23.36 | Improved |
| apply integration/decomposition as proxy task to iCAN | iCAN · HICO-DET Default Full | mAP: 14.84 -> 18.98 | Improved |
| use HICO-DET fine-tuned (DRG) detector instead of COCO detector | IDN · HICO-DET Default Full | mAP: 23.36 -> 26.29 | 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.
- VSGNet — VSGNet: Spatial Attention Network for Detecting Human Object Interactions Using Graph Convolutions (CVPR 2020) · better on 2 of 2 shared results
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
- Detecting Any Human-Object Interaction Relationship: Universal HOI Detector with Spatial Prompt Learning on Foundation Models (NeurIPS 2023 · similarity 0.90)
- End-to-End Human Object Interaction Detection With HOI Transformer (CVPR 2021 · similarity 0.90)
- Neural-Logic Human-Object Interaction Detection (NeurIPS 2023 · similarity 0.90)
- Exploring Structure-Aware Transformer Over Interaction Proposals for Human-Object Interaction Detection (CVPR 2022 · similarity 0.90)
- Open-World Human-Object Interaction Detection via Multi-modal Prompts (CVPR 2024 · similarity 0.88)
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 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.
