SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data

9 Dec 2020  ·  Shaoli Huang, Xinchao Wang, DaCheng Tao ·

Data mixing augmentation has proved effective in training deep models. Recent methods mix labels mainly based on the mixture proportion of image pixels. As the main discriminative information of a fine-grained image usually resides in subtle regions, methods along this line are prone to heavy label noise in fine-grained recognition. We propose in this paper a novel scheme, termed as Semantically Proportional Mixing (SnapMix), which 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, and allows for asymmetric mixing operations and ensures semantic correspondence between synthetic images and target labels. Experiments show that our method consistently outperforms existing mixed-based approaches on various datasets and under different network depths. Furthermore, by incorporating the mid-level features, the proposed SnapMix achieves top-level performance, demonstrating its potential to serve as a solid baseline for fine-grained recognition. Our code is available at

PDF Abstract


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Fine-Grained Image Classification CUB-200-2011 SnapMix Accuracy 89.58% # 30


No methods listed for this paper. Add relevant methods here