Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition

This paper does not attempt to design a state-of-the-art method for visual recognition but investigates a more efficient way to make use of convolutions to encode spatial features. By comparing the design principles of the recent convolutional neural networks ConvNets) and Vision Transformers, we pr...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Hou, Qibin, Cheng-Ze Lu, Ming-Ming, Cheng, Feng, Jiashi
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 22.11.2022
Subjects
Artificial neural networks
Image segmentation
Modulation
Object recognition
Vision
Online AccessGet full text

Cover

More Information
Summary:This paper does not attempt to design a state-of-the-art method for visual recognition but investigates a more efficient way to make use of convolutions to encode spatial features. By comparing the design principles of the recent convolutional neural networks ConvNets) and Vision Transformers, we propose to simplify the self-attention by leveraging a convolutional modulation operation. We show that such a simple approach can better take advantage of the large kernels (>=7x7) nested in convolutional layers. We build a family of hierarchical ConvNets using the proposed convolutional modulation, termed Conv2Former. Our network is simple and easy to follow. Experiments show that our Conv2Former outperforms existent popular ConvNets and vision Transformers, like Swin Transformer and ConvNeXt in all ImageNet classification, COCO object detection and ADE20k semantic segmentation.
ISSN:2331-8422