Pure MLP Methods And What They Are Suitable For

Previously I have spent some time researching on pure MLP-based segmentation methods. I did find some amazing work that play all kinds of tricks on MLP, but still, i give up on segmentation MLPs in the end. Generally speaking, MLPs are not suitable for dense prediction tasks unless some groundbreaking new architecture are proposed. How it started After Vision Transformers storm the Computer Vision community, people have paid more and more attention on finding ways of capturing long-range depende...

Submit