SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data

Authors

  • Shaoli Huang The University of Sydney
  • Xinchao Wang Stevens Institute of Technology
  • Dacheng Tao The University of Sydney

Keywords:

Object Detection & Categorization, Learning & Optimization for CV, Classification and Regression

Abstract

Data mixing augmentation has proved effective in training deep models. Recent methods mix labels mainly according to the mixture proportion of image pixels. Due to the major discriminative information of a fine-grained image usually resides in subtle regions, these methods tend to introduce heavy label noise in fine-grained recognition. We propose Semantically Proportional Mixing (SnapMix) that exploits class activation map (CAM) to lessen the label noise in augmenting fine-grained data. SnapMix generates the target label for a mixed image by estimating its intrinsic semantic composition. This strategy can adapt to asymmetric mixing operations and ensure semantic correspondence between synthetic images and target labels. Experiments show that our method consistently outperforms existing mixed-based approaches regardless of different datasets or network depths. Further, by incorporating the mid-level features, the proposed SnapMix achieves top-level performance, demonstrating its potential to serve as a strong baseline for fine-grained recognition.

Downloads

Published

2021-05-18

How to Cite

Huang, S., Wang, X., & Tao, D. (2021). SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data. Proceedings of the AAAI Conference on Artificial Intelligence, 35(2), 1628-1636. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/16255

Issue

Section

AAAI Technical Track on Computer Vision I