Asymmetric patch sampling for contrastive learning

Chengchao Shen, Jianzhong Chen, Shu Wang, Hulin Kuang, Jin Liu, Jianxin Wang

Abstract: Highlights•Excessive appearance similarities hurt the performance of contrastive learning.•We propose a novel asymmetric patch sampling strategy for contrastive learning.•Our method reduces the appearance similarities but retains the image semantics.•Our method addresses non-unique target issue of masked image modeling methods.•Experiments show that our method achieves state-of-the-art performance.