Domain generalization for image classification with dynamic decision boundary

Zhiming Cheng, Mingxia Liu, Defu Yang, Zhidong Zhao, Chenggang Yan, Shuai Wang

Abstract: Highlights•We proposed an innovative dynamic decision boundary-based domain generalization.•We designed a data augmentation method called USM to address overfitting issues.•We propose a PMP module to dynamically adjust the model’s decision boundary.•Experiment results show that our method achieves state-of-the-art performance.

External IDs:doi:10.1016/j.patcog.2025.111678