Research map / Detection and segmentation / Open-set and few-shot detection
Consistency Learning via Decoding Path Augmentation for Transformers in Human Object Interaction Detection
Jihwan Park, SeungJun Lee, Hwan Heo, Hyeong Kyu Choi, Hyunwoo J. Kim
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 cross-path consistency training to QPIC | QPIC · V-COCO test, AProle1 | mAP: 62.2* -> 63.1 | Improved |
| Add cross-path consistency training to QPIC | QPIC · V-COCO test, AProle2 | mAP: 64.5* -> 65.4 | Improved |
| Add cross-path consistency training to HOTR | HOTR · V-COCO test, AProle1 | mAP: 59.8* -> 61.6 | Improved |
| Add cross-path consistency training to HOTR | HOTR · V-COCO test, AProle2 | mAP: 64.9* -> 66.8 | Improved |
| Add cross-path consistency training to QPIC | QPIC · HICO-DET, Full | mAP: 29.07 -> 29.63 | Improved |
| Add cross-path consistency training to QPIC | QPIC · HICO-DET, Rare | mAP: 21.85 -> 23.14 | Improved |
| Add cross-path consistency training to QPIC | QPIC · HICO-DET, Non-Rare | mAP: 31.23 -> 31.57 | Improved |
| Add cross-path consistency training to HOTR | HOTR · HICO-DET, Full | mAP: 25.10 -> 26.16 | Improved |
| Add cross-path consistency training to HOTR | HOTR · HICO-DET, Rare | mAP: 17.34 -> 22.84 | Improved |
| Add cross-path consistency training to HOTR | HOTR · HICO-DET, Non-Rare | mAP: 27.42 -> 27.15 | Hurt |
| Add decoder sharing across paths while retaining CPC (ablation) | QPIC · V-COCO test, AProle1, path P1 | mAP: 62.4 -> 63.1 | Improved |
| Add decoder sharing across paths while retaining CPC (ablation) | HOTR · V-COCO test, AProle1, path P1 | mAP: 61.2 -> 61.6 | 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.
- HOTR — HOTR: End-to-End Human-Object Interaction Detection With Transformers (CVPR 2021) · better on 5 of 5 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
- Human-Object Interaction Detection via Disentangled Transformer (CVPR 2022 · similarity 0.91)
- End-to-End Human Object Interaction Detection With HOI Transformer (CVPR 2021 · similarity 0.91)
- Exploring Structure-Aware Transformer Over Interaction Proposals for Human-Object Interaction Detection (CVPR 2022 · similarity 0.90)
- LINK: Learning Instance-level Knowledge from Vision-Language Models for Human-Object Interaction Detection (ICLR 2026 · similarity 0.90)
- Efficient Two-Stage Detection of Human-Object Interactions With a Novel Unary-Pairwise Transformer (CVPR 2022 · similarity 0.90)
- Dual-Prior Augmented Decoding Network for Long Tail Distribution in HOI Detection (AAAI 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.
