4 papers
From Semantics to Pixels: Coarse-to-Fine Masked Autoencoders for Hierarchical Visual Understanding
Wenzhao Xiang, Yue Wu, Hongyang Yu +3
Self-supervised visual pre-training methods face an inherent tension: contrastive learning (CL) captures global semantics but loses fine-grained detail, while masked image modeling…
Wavelet-Driven Masked Image Modeling: A Path to Efficient Visual Representation
Wenzhao Xiang, Chang Liu, Hongyang Yu +1
Masked Image Modeling (MIM) has garnered significant attention in self-supervised learning, thanks to its impressive capacity to learn scalable visual representations tailored for…
AEMIM: Adversarial Examples Meet Masked Image Modeling
Wenzhao Xiang, Chang Liu, Hang Su +1
Masked image modeling (MIM) has gained significant traction for its remarkable prowess in representation learning. As an alternative to the traditional approach, the reconstruction…
Improving Model Generalization by On-manifold Adversarial Augmentation in the Frequency Domain
Chang Liu, Wenzhao Xiang, Yuan He +3
Deep neural networks (DNNs) may suffer from significantly degenerated performance when the training and test data are of different underlying distributions. Despite the importance…