arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.16964cs.CV
收藏

基于因果表示的视线估计领域泛化

Causal Representation-Based Domain Generalization on Gaze Estimation

Gachon University(嘉泉大学);VisualCamp(视觉营地);Sungkyunkwan University(成均馆大学)

Younghan Kim, Kangryun Moon, Yongjun Park, Yonggyu Kim

AI总结:

本文提出CauGE框架,通过对抗训练与惩罚项提取领域不变特征,并利用注意力层提升特征充分性,从而消除虚假相关影响,在视线估计领域泛化基准上达到最先进性能。

摘要:

包含每个受试者视线信息的大规模数据集的可用性显著提高了视线估计的准确性。然而,领域之间的差异严重影响了为特定领域显式训练的模型的性能。在本文中,我们提出了基于因果表示的视线估计领域泛化(CauGE)框架,该框架基于与领域差异相一致的因果机制一般原理设计。我们采用对抗训练方式和一个额外的惩罚项来提取领域不变特征。在提取特征之后,我们放置注意力层,使特征足以推断实际视线。通过利用这些模块,CauGE确保神经网络从满足因果机制一般原理的表示中学习。由此,CauGE通过提取领域不变特征实现跨领域泛化,虚假相关性无法影响模型。我们的方法在视线估计基准的领域泛化任务中取得了最先进的性能。

英文摘要:

The availability of extensive datasets containing gaze information for each subject has significantly enhanced gaze estimation accuracy. However, the discrepancy between domains severely affects a model's performance explicitly trained for a particular domain. In this paper, we propose the Causal Representation-Based Domain Generalization on Gaze Estimation (CauGE) framework designed based on the general principle of causal mechanisms, which is consistent with the domain difference. We employ an adversarial training manner and an additional penalizing term to extract domain-invariant features. After extracting features, we position the attention layer to make features sufficient for inferring the actual gaze. By leveraging these modules, CauGE ensures that the neural networks learn from representations that meet the causal mechanisms' general principles. By this, CauGE generalizes across domains by extracting domain-invariant features, and spurious correlations cannot influence the model. Our method achieves state-of-the-art performance in the domain generalization on gaze estimation benchmark.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑