HackerPapers

arXiv:2201.03545cs.CV

A ConvNet for the 2020s

Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie

Abstract

The "Roaring 20s" of visual recognition began with the introduction of Vision Transformers (ViTs), which quickly superseded ConvNets as the state-of-the-art image classification model. A vanilla ViT, on the other hand, faces difficulties when applied to general computer vision tasks such as object detection and semantic segmentation. It is the hierarchical Transformers (e.g., Swin Transformers) that reintroduced several ConvNet priors, making Transformers practically viable as a generic vision backbone and demonstrating remarkable performance on a wide variety of vision tasks. However, the effectiveness of such hybrid approaches is still largely credited to the intrinsic superiority of Transformers, rather than the inherent inductive biases of convolutions. In this work, we reexamine the design spaces and test the limits of what a pure ConvNet can achieve. We gradually "modernize" a standard ResNet toward the design of a vision Transformer, and discover several key components that contribute to the performance difference along the way. The outcome of this exploration is a family of pure ConvNet models dubbed ConvNeXt. Constructed entirely from standard ConvNet modules, ConvNeXts compete favorably with Transformers in terms of accuracy and scalability, achieving 87.8% ImageNet top-1 accuracy and outperforming Swin Transformers on COCO detection and ADE20K segmentation, while maintaining the simplicity and efficiency of standard ConvNets.

arXiv comments
CVPR 2022; Code: https://github.com/facebookresearch/ConvNeXt
Journal reference
DOI
Subjects
cs.CVComputer Vision and Pattern Recognition

On Hacker News

Linked by 6 identifiable commenters in 6 comments across 5 threads, from Apr 2022 to Aug 2024. Found in 3 HN submissions from Jan 2022 to May 2024. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments5 threads

  1. What's Going on in Machine Learning? Some Minimal Models 1 comment

    • Perhaps; in a lot of cases the architecture barely matters. Transformers took a lot of extra tricks to get working well; the ConvNext paper showed that applying those same tricks to convolutional networks can fully close the gap. https://arxiv.org/abs/2201.03545

      sdenton4View on HNParent context

  2. The Path to StyleGan2 – Implementing the Progressive Growing GAN 1 comment

    • > The rule in ML There's definitely attempts to revive things (in the general sense, not just GANs), but most successes appear to come from large labs dedicating equal computational resources to the older models and often by changing names. This can make things more confusing…

      godelskiView on HNParent context

  3. Citation cartels help mathematicians-and their universities-climb the rankings 1 comment

    • Academic papers often include an introduction which summarises the context in which the paper is written. Consider https://arxiv.org/pdf/2201.03545v2.pdf where the paper says "For many decades, this has been the default use of ConvNets, generally on limited object categories…

      michaeltView on HNParent context

  4. Swin Transformer: Hierarchical Vision Transformer Using Shifted Windows 1 comment

    • Researchers love the ViT and all its varieties, but a reminder for those following at home that ConvNets still work and scale fine, depending on your requirements. For example, in "A ConvNet for the 2020s"[1], the authors are able to scale up ConvNets to the sizes of ViTs. From…

      nharadaView on HNParent context

  5. Transformers Are All You Need 2 comments

    • The performance advantages of transformer architectures have little to do with "transformers" themselves, but rather other architecture changes that were made along the way. They can be matched or beat in performance by tweaked conventional CNNs: https://arxiv.org/abs/2201.03545

      riotnrrdView on HNParent context

    • > ViT models are outperforming CNNs in terms of computational efficiency and accuracy, achieving highly competitive performance in tasks like image classification, object detection, and semantic image segmentation. Since then, this has been show to be untrue. Using more modern…

      mrintellectualView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.