解决人物交互检测中的相似类别干扰问题，提升识别精度。
No More Sibling Rivalry: Debiasing Human-Object Interaction Detection
- 提出两种去偏学习策略，分别从输入和输出端缓解相似交互对的竞争。
- 在HICO-Det数据集上达到9.18%的mAP提升，超过当前最优模型3.59%。
- 适合关注图像中细粒度交互理解与模型公平性的研究者。
检测变压器已被用于人-物交互（HOI）检测，提升了图像中人-动作-物三元组的定位与识别能力。尽管进展显著，本研究发现一个关键问题——“有毒兄弟”偏差，即大量相似但不同的HOI三元组在输入和输出端相互干扰甚至竞争，阻碍交互解码器的学习。该偏差源于兄弟类别间高度混淆，相似性越高反而精度越低，因一方收益导致另一方损失。为此，我们提出两种新型去偏学习目标：‘对比-校准’和‘合并-拆分’，分别针对输入和输出视角。前者通过采样相似但错误的三元组并利用强位置先验重构为正确形式；后者先学习兄弟类别间的共享特征以区分其他组，再显式细化组内差异以保持独特性。实验表明，我们在多种设置下显著超越基线（在HICO-Det上提升9.18% mAP）和现有最优方法（提升3.59% mAP）。
原文摘要 · Abstract (English)
Detection transformers have been applied to human-object interaction (HOI) detection, enhancing the localization and recognition of human-action-object triplets in images. Despite remarkable progress, this study identifies a critical issue-"Toxic Siblings" bias-which hinders the interaction decoder's learning, as numerous similar yet distinct HOI triplets interfere with and even compete against each other both input side and output side to the interaction decoder. This bias arises from high confusion among sibling triplets/categories, where increased similarity paradoxically reduces precision, as one's gain comes at the expense of its toxic sibling's decline. To address this, we propose two novel debiasing learning objectives-"contrastive-then-calibration" and "merge-then-split"-targeting the input and output perspectives, respectively. The former samples sibling-like incorrect HOI triplets and reconstructs them into correct ones, guided by strong positional priors. The latter first learns shared features among sibling categories to distinguish them from other groups, then explicitly refines intra-group differentiation to preserve uniqueness. Experiments show that we significantly outperform both the baseline (+9.18% mAP on HICO-Det) and the state-of-the-art (+3.59% mAP) across various settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。