MC-MLP:Multiple Coordinate Frames in all-MLP Architecture for Vision

TL;DR

MC-MLP introduces a novel all-MLP architecture for vision that employs multiple coordinate frames through orthogonal transforms, enhancing feature learning and outperforming existing MLP models in image classification.

Contribution

The paper proposes MC-MLP, a new MLP backbone that uses multiple coordinate frames via orthogonal transforms to improve learning capacity in vision tasks.

Findings

MC-MLP outperforms most existing MLPs in image classification.

The model achieves better performance at the same parameter level.

Orthogonal transforms enable learning across different coordinate frames.

Abstract

In deep learning, Multi-Layer Perceptrons (MLPs) have once again garnered attention from researchers. This paper introduces MC-MLP, a general MLP-like backbone for computer vision that is composed of a series of fully-connected (FC) layers. In MC-MLP, we propose that the same semantic information has varying levels of difficulty in learning, depending on the coordinate frame of features. To address this, we perform an orthogonal transform on the feature information, equivalent to changing the coordinate frame of features. Through this design, MC-MLP is equipped with multi-coordinate frame receptive fields and the ability to learn information across different coordinate frames. Experiments demonstrate that MC-MLP outperforms most MLPs in image classification tasks, achieving better performance at the same parameter level. The code will be available at: https://github.com/ZZM11/MC-MLP.