An Image Patch is a Wave: Quantum Inspired Vision MLP
Preprint
Abstract
In the field of computer vision, recent works show that a pure MLP architecture mainly stacked by fully-connected layers can achieve competing performance with CNN and transformer. An input image of vision MLP is usually split into multiple tokens (patches), while the existing MLP models directly aggregate them with fixed weights, neglecting the varying semantic information of tokens from different images. To dynamically aggregate tokens, we propose to represent each token as a wave function with two parts, amplitude and phase. Amplitude is the original feature and the phase term is a complex value changing according to the semantic contents of input images. Introducing the phase term can dynamically modulate the relationship between tokens and fixed weights in MLP. Based on the wave-like token representation, we establish a novel Wave-MLP architecture for vision tasks. Extensive experiments demonstrate that the proposed Wave-MLP is superior to the state-of-the-art MLP architectures on various vision tasks such as image classification, object detection and semantic segmentation. The source code will be available at https://github.com/huawei-noah/CV-Backbones/tree/master/wavemlp_pytorch and https://gitee.com/mindspore/models/tree/master/research/cv/wave_mlp.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- CMT: Convolutional Neural Networks Meet Vision Transformers2022
- Patch Slimming for Efficient Vision Transformers2022
- Hire-MLP: Vision MLP via Hierarchical Rearrangement2022
- Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions2021
- Rethinking Spatial Dimensions of Vision Transformers2021
- Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet2021
- Semantic Understanding of Scenes Through the ADE20K DatasetInternational Journal of Computer Vision, 2018
- A straightforward introduction to continuous quantum measurementContemporary Physics, 2006
- Wave–particle duality of C60 moleculesNature, 1999
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998