ResNeSt: Split-Attention Networks

Hang Zhang, Chongruo Wu, Zhongyue Zhang, Yi Zhu, Haibin Lin +7 more
4/19/2020
cs.CV

Official code: zhanghang1989/ResNeSt (3,260 stars, Apache-2.0, PyTorch, last commit 2026-09-11).

Abstract

It is well known that featuremap attention and multi-path representation are important for visual recognition. In this paper, we present a modularized architecture, which applies the channel-wise attention on different network branches to leverage their success in capturing cross-feature interactions and learning diverse representations. Our design results in a simple and unified computation block, which can be parameterized using only a few variables. Our model, named ResNeSt, outperforms EfficientNet in accuracy and latency trade-off on image classification. In addition, ResNeSt has achieved superior transfer learning results on several public benchmarks serving as the backbone, and has been adopted by the winning entries of COCO-LVIS challenge. The source code for complete system and pretrained models are publicly available.

View on arXivView PDF

Code Implementations(1)

zhanghang1989/ResNeStOfficial100%
PyTorchApache-2.0

ResNeSt: Split-Attention Networks

3,260494PythonMar 15, 20201 months agoApache-2.0
deep-learningdetectron-modelspytorchresnestresnet+1 more

Similar Papers

DIANet: Dense-and-Implicit Attention Network

85% match1 repo1022019

EMBANet: A Flexible Efffcient Multi-branch Attention Network

85% match2024

End-to-End Multi-Task Learning with Attention

84% match1 repo7372018

MLP-Mixer: An all-MLP Architecture for Vision

84% match2 repos36.9k2021

Vision Backbone Enhancement via Multi-Stage Cross-Scale Attention

83% match2023

Dyn-Adapter: Towards Disentangled Representation for Efficient Visual Recognition

82% match2024

Cite this paper

@article{zhang2020resnest,
  title  = {ResNeSt: Split-Attention Networks},
  author = {Hang Zhang and Chongruo Wu and Zhongyue Zhang and Yi Zhu and Haibin Lin and Zhi Zhang and Yue Sun and Tong He and Jonas Mueller and R. Manmatha and Mu Li and Alexander Smola},
  year   = {2020},
  eprint = {2004.08955},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2004.08955}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2004.08955.svg)](https://codeofpaper.com/paper/2004.08955)

Discussion