Panoptic Scene Graph Generation (PSG) parses objects and predicts their relationships (predicate) to connect human language and visual scenes. However, different language preferences of annotators and semantic overlaps between predicates lead to biased predicate annotations in the dataset, i.e. different predicates for same object pairs. Biased predicate annotations make PSG models struggle in constructing a clear decision plane among predicates, which greatly hinders the real application of PSG models. To address the intrinsic bias above, we propose a novel framework named ADTrans to adaptively transfer biased predicate annotations to informative and unified ones. To promise consistency and accuracy during the transfer process, we propose to measure the invariance of representations in each predicate class, and learn unbiased prototypes of predicates with different intensities. Meanwhile, we continuously measure the distribution changes between each presentation and its prototype, and constantly screen potential biased data. Finally, with the unbiased predicate-prototype representation embedding space, biased annotations are easily identified. Experiments show that ADTrans significantly improves the performance of benchmark models, achieving a new state-of-the-art performance, and shows great generalization and effectiveness on multiple datasets.

翻译：全景场景图生成（Panoptic Scene Graph Generation, PSG）旨在解析对象并预测它们之间的关系（谓词），以连接人类语言与视觉场景。然而，标注者不同的语言偏好以及谓词间的语义重叠，导致数据集中出现有偏的谓词标注，即相同对象对对应不同谓词。有偏的谓词标注使得PSG模型难以在谓词间构建清晰的决策平面，严重阻碍了PSG模型的实际应用。为克服上述内在偏差，本文提出名为ADTrans的新型框架，自适应地将有偏谓词标注转化为信息丰富且统一的标注。为确保转换过程中的一致性与准确性，我们提出度量每个谓词类别中表征的不变性，并学习不同强度的谓词无偏原型。同时，持续度量每个表征与其原型之间的分布变化，不断筛选潜在有偏数据。最终，借助无偏的谓词-原型表征嵌入空间，有偏标注可被轻松识别。实验表明，ADTrans显著提升了基准模型的性能，达到新的最优水平，并在多个数据集上展现出良好的泛化性和有效性。