# Shape-Texture Debiased Neural Network Training

arXiv: https://arxiv.org/abs/2010.05981 · HTML: https://mlsanity.com/paper/2010.05981/

## Key facts

- Cited by 76 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Oct 2020.

## Built on by (most cited first, 30 of 76)

- [Towards Robust Vision Transformer](https://mlsanity.com/paper/2105.07926/) (2021)
- [A Comprehensive Survey of Image Augmentation Techniques for Deep Learning](https://mlsanity.com/paper/2205.01491/) (2022)
- [AugMax: Adversarial Composition of Random Augmentations for Robust Training](https://mlsanity.com/paper/2110.13771/) (2021)
- [MedViT: A Robust Vision Transformer for Generalized Medical Image Classification](https://mlsanity.com/paper/2302.09462/) (2023)
- [RobustART: Benchmarking Robustness on Architecture Design and Training Techniques](https://mlsanity.com/paper/2109.05211/) (2021)
- [Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?](https://mlsanity.com/paper/2106.02890/) (2021)
- [Human alignment of neural network representations](https://mlsanity.com/paper/2211.01201/) (2022)
- [Content-based Unrestricted Adversarial Attack](https://mlsanity.com/paper/2305.10665/) (2023)
- [Robust Contrastive Learning Using Negative Samples with Diminished Semantics](https://mlsanity.com/paper/2110.14189/) (2021)
- [Harnessing Out-Of-Distribution Examples via Augmenting Content and Style](https://mlsanity.com/paper/2207.03162/) (2022)
- [Robust and Accurate Object Detection via Adversarial Learning](https://mlsanity.com/paper/2103.13886/) (2021)
- [Unbiased Supervised Contrastive Learning](https://mlsanity.com/paper/2211.05568/) (2022)
- [Bias Mimicking: A Simple Sampling Approach for Bias Mitigation](https://mlsanity.com/paper/2209.15605/) (2022)
- [ImageNet-E: Benchmarking Neural Network Robustness via Attribute Editing](https://mlsanity.com/paper/2303.17096/) (2023)
- [Taxonomy of Machine Learning Safety: A Survey and Primer](https://mlsanity.com/paper/2106.04823/) (2021)
- [Removing Batch Normalization Boosts Adversarial Training](https://mlsanity.com/paper/2207.01156/) (2022)
- [On the Effect of Pruning on Adversarial Robustness](https://mlsanity.com/paper/2108.04890/) (2021)
- [Enhance the Visual Representation via Discrete Adversarial Training](https://mlsanity.com/paper/2209.07735/) (2022)
- [Can We Talk Models Into Seeing the World Differently?](https://mlsanity.com/paper/2403.09193/) (2024)
- [Can Biases in ImageNet Models Explain Generalization?](https://mlsanity.com/paper/2404.01509/) (2024)
- [Not Just Pretty Pictures: Toward Interventional Data Augmentation Using Text-to-Image Generators](https://mlsanity.com/paper/2212.11237/) (2022)
- [A Game-Theoretic Taxonomy of Visual Concepts in DNNs](https://mlsanity.com/paper/2106.10938/) (2021)
- [Contributions of Shape, Texture, and Color in Visual Recognition](https://mlsanity.com/paper/2207.09510/) (2022)
- [On-target Adaptation](https://mlsanity.com/paper/2109.01087/) (2021)
- [DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery](https://mlsanity.com/paper/2504.04804/) (2025)
- [Exploring Corruption Robustness: Inductive Biases in Vision Transformers and MLP-Mixers](https://mlsanity.com/paper/2106.13122/) (2021)
- [Controlling Directions Orthogonal to a Classifier](https://mlsanity.com/paper/2201.11259/) (2022)
- [ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression](https://mlsanity.com/paper/2509.20234/) (2025)
- [Combining Diverse Feature Priors](https://mlsanity.com/paper/2110.08220/) (2021)
- [Convolutional Neural Networks Rarely Learn Shape for Semantic Segmentation](https://mlsanity.com/paper/2305.06568/) (2023)

## Abstract (arXiv)

Shape and texture are two prominent and complementary cues for recognizing objects. Nonetheless, Convolutional Neural Networks are often biased towards either texture or shape, depending on the training dataset. Our ablation shows that such bias degenerates model performance. Motivated by this observation, we develop a simple algorithm for shape-texture debiased learning. To prevent models from exclusively attending on a single cue in representation learning, we augment training data with images with conflicting shape and texture information (eg, an image of chimpanzee shape but with lemon texture) and, most importantly, provide the corresponding supervisions from shape and texture simultaneously. Experiments show that our method successfully improves model performance on several image recognition benchmarks and adversarial robustness. For example, by training on ImageNet, it helps ResNet-152 achieve substantial improvements on ImageNet (+1.2%), ImageNet-A (+5.2%), ImageNet-C (+8.3%) and Stylized-ImageNet (+11.1%), and on defending against FGSM adversarial attacker on ImageNet (+14.4%). Our method also claims to be compatible with other advanced data augmentation strategies, eg, Mixup, and CutMix. The code is available here: https://github.com/LiYingwei/ShapeTextureDebiasedTraining.