2022-05-30 更新
X-ViT: High Performance Linear Vision Transformer without Softmax
Authors:Jeonggeun Song, Heung-Chang Lee
Vision transformers have become one of the most important models for computer vision tasks. Although they outperform prior works, they require heavy computational resources on a scale that is quadratic to the number of tokens, $N$. This is a major drawback of the traditional self-attention (SA) algorithm. Here, we propose the X-ViT, ViT with a novel SA mechanism that has linear complexity. The main approach of this work is to eliminate nonlinearity from the original SA. We factorize the matrix multiplication of the SA mechanism without complicated linear approximation. By modifying only a few lines of code from the original SA, the proposed models outperform most transformer-based models on image classification and dense prediction tasks on most capacity regimes.
PDF
论文截图
Architecture-Agnostic Masked Image Modeling — From ViT back to CNN
Authors:Siyuan Li, Di Wu, Fang Wu, Zelin Zang, Kai Wang, Lei Shang, Baigui Sun, Hao Li, Stan. Z. Li
Masked image modeling (MIM), an emerging self-supervised pre-training method, has shown impressive success across numerous downstream vision tasks with Vision transformers (ViT). Its underlying idea is simple: a portion of the input image is randomly masked out and then reconstructed via the pre-text task. However, why MIM works well is not well explained, and previous studies insist that MIM primarily works for the Transformer family but is incompatible with CNNs. In this paper, we first study interactions among patches to understand what knowledge is learned and how it is acquired via the MIM task. We observe that MIM essentially teaches the model to learn better middle-level interactions among patches and extract more generalized features. Based on this fact, we propose an Architecture-Agnostic Masked Image Modeling framework (A$^2$MIM), which is compatible with not only Transformers but also CNNs in a unified way. Extensive experiments on popular benchmarks show that our A$^2$MIM learns better representations and endows the backbone model with the stronger capability to transfer to various downstream tasks for both Transformers and CNNs.
PDF A preprint version. The source code will be released in https://github.com/Westlake-AI/openmixup
论文截图