ConvMLP: Hierarchical Convolutional MLPs for Vision

Abstract

MLP-based architectures, which consist of a sequence of consecutive multi-layer perceptron blocks, have recently been found to reach comparable results to convolutional and transformer-based methods. However, most adopt spatial MLPs which take fixed dimension inputs, therefore making it difficult to apply them to downstream tasks, such as object detection and semantic segmentation. Moreover, single-stage designs further limit performance in other computer vision tasks and fully connected layers bear heavy computation. To tackle these problems, we propose ConvMLP: a hierarchical Convolutional MLP for visual recognition, which is a light-weight, stage-wise, co-design of convolution layers, and MLPs. In particular, ConvMLP-S achieves 76.8% top-1 accuracy on ImageNet-1k with 9M parameters and 2.4G MACs (15% and 19% of MLP-Mixer-B/16, respectively). Experiments on object detection and semantic segmentation further show that visual representation learned by ConvMLP can be seamlessly transferred and achieve competitive results with fewer parameters. Our code and pre-trained models are publicly available at https://github.com/SHI-Labs/Convolutional-MLPs.

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• MLP-Mixer: An all-MLP Architecture for Vision
• MMDetection: Open MMLab Detection Toolbox and Benchmark
• Shuffle Transformer: Rethinking Spatial Shuffle for Vision Transformer
• Pay Attention to MLPs
• S${}^2$-MLP: Spatial-Shift MLP Architecture for Vision
• Vision Permutator: A Permutable MLP-Like Architecture for Visual Recognition