Research map / Detection and segmentation / Open-set and few-shot detection
Polysemy Deciphering Network for Human-Object Interaction Detection
Xubin Zhong, Changxing Ding, Xian Qu, Dacheng Tao
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 Polysemy-Aware Modal Fusion (PAMF) to the baseline (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 1.29% and 1.36% mAP in DT and KO modes | Improved |
| add Language Prior-based Feature Augmentation (LPFA) (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 0.52% and 0.21% mAP in DT and KO modes | Improved |
| add Language Prior-guided Channel Attention (LPCA) (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 1.33% and 0.21% in DT and KO modes | Improved |
| add clustering-based object-specific (CSP) classifiers instead of shared (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 1.06% and 2.13% mAP in DT and KO modes | Improved |
| add INet (suppress non-interacting pairs) (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 21.77 -> 22.37 (DT); 26.98 -> 26.86 (KO) | Mixed |
| add object word embedding to the language prior (verb + object vs verb only) (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 19.98 -> 22.37 (DT); 24.99 -> 26.86 (KO) | Improved |
| add auxiliary supervision S_au to LPCA (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 19.41 -> 22.37 (DT); 23.32 -> 26.86 (KO) | Improved |
| use LPCA instead of plain channel attention (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 19.97 -> 22.37 (DT); 24.11 -> 26.86 (KO) | Improved |
| use LPCA instead of FiLM (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 20.90 -> 22.37 (DT); 25.29 -> 26.86 (KO) | Improved |
| add C_att plain CA refinement to LPCA (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 21.99 -> 22.37 (DT); 26.34 -> 26.86 (KO) | Improved |
| use CSP classifiers instead of shared (SH) classifiers (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT/KO: 21.06 -> 22.37 (DT); 24.83 -> 26.86 (KO) | Improved |
| use object-shared (SH) instead of object-specific (SP) classifiers on rare categories (ablation) | PD-Net · HICO-DET (600 HOI, Full) | mAP DT rare: 15.03 -> 16.45 (DT rare); 17.83 -> 19.89 (KO rare) | 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 1 of 1 shared result
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
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 2024 · similarity 0.91)
- Discovering Syntactic Interaction Clues for Human-Object Interaction Detection (CVPR 2024 · similarity 0.90)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.90)
- Toward Open-Set Human Object Interaction Detection (AAAI 2024 · similarity 0.89)
- Open-World Human-Object Interaction Detection via Multi-modal Prompts (CVPR 2024 · similarity 0.89)
- Human-Object Interaction Detection Collaborated with Large Relation-driven Diffusion Models (NeurIPS 2024 · similarity 0.89)
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.
