MCMAE: Masked Convolution Meets Masked Autoencoders

Vision Transformers (ViT) become widely-adopted architectures for various vision tasks. Masked auto-encoding [2, 1, 28, 55] for feature pretraining and multi-scale hybrid convolution-transformer architectures [12, 21, 49, 34, 57] can further unleash the potentials of ViT, leading to state-of-the-art performances on image classification, detection and semantic segmentation. In this paper, our MCMAE framework demonstrates that multi-scale hybrid convolution-transformer can learn more discriminative representations via the mask auto-encoding scheme. However, directly using the original masking strategy leads to the heavy computational cost and pretraining-finetuning discrepancy. To tackle the issue, we adopt the masked convolution to prevent information leakage in the convolution blocks. A simple block-wise masking strategy is proposed to ensure computational efficiency. We also propose to more directly supervise the multi-scale features of the encoder to boost multi-scale features. MCMAE-Base improves ImageNet-1K finetuning accuracy by 1.4% compared with MAE-Base. On object detection, MCMAE-Base finetuned for only 25 epochs surpasses MAE-Base fined-tuned for 100 epochs by 2.9% AP box and 2.2% AP mask respectively. Code and pretrained models are available at https://github.com/Alpha-VL/ConvMAE .