paper-with-me

Papers

Err on the Side of Texture: Texture Bias on Real Data

2024-12-13 · Blaine Hoak, Ryan Sheatsley, Patrick McDaniel

Bias significantly undermines both the accuracy and trustworthiness of machine learning models. To date, one of the strongest biases observed in image classification models is texture bias-where models overly rely on texture information rather than shape information. Yet, existing approaches for measuring and mitigating texture bias have not been able to capture how textures impact model robustness in real-world settings. In this work, we introduce the Texture Association Value (TAV), a novel metric that quantifies how strongly models rely on the presence of specific textures when classifying objects. Leveraging TAV, we demonstrate that model accuracy and robustness are heavily influenced by texture. Our results show that texture bias explains the existence of natural adversarial examples, where over 90% of these samples contain textures that are misaligned with the learned texture of their true label, resulting in confident mispredictions.

📄 PDF Abstract BibTeX arXiv:2412.10597

Code (1)

blainehoak/err-on-textures 공식 구현 pytorch

Tasks

image-classificationImage Classification

Similar Papers 제목 키워드 기반

Content Preserving Image Translation with Texture Co-occurrence and Spatial Self-Similarity for Texture Debiasing and Domain Adaptation

2021-10-15 · Myeongkyun Kang, Dongkyu Won, Miguel Luna, Philip Chikontwe 외

Models trained on datasets with texture bias usually perform poorly on out-of-distribution samples since biased representations are embedded into the model. Recently, various image translation and debiasing methods have …

Domain AdaptationTranslation

BiasBed -- Rigorous Texture Bias Evaluation

2022-11-23 · Nikolai Kalischek, Rodrigo C. Daudt, Torben Peters, Reinhard Furrer 외

The well-documented presence of texture bias in modern convolutional neural networks has led to a plethora of algorithms that promote an emphasis on shape cues, often to support generalization to new domains. Yet, common…

Model Selection

BiasBed - Rigorous Texture Bias Evaluation

2023-01-01 · CVPR 2023 1 · Nikolai Kalischek, Rodrigo Caye Daudt, Torben Peters, Reinhard Furrer 외

The well-documented presence of texture bias in modern convolutional neural networks has led to a plethora of algorithms that promote an emphasis on shape cues, often to support generalization to new domains. Yet, co…

Model Selection

TextureSAM: Towards a Texture Aware Foundation Model for Segmentation

2025-05-22 · Inbal Cohen, Boaz Meivar, Peihan Tu, Shai Avidan 외

Segment Anything Models (SAM) have achieved remarkable success in object segmentation tasks across diverse datasets. However, these models are predominantly trained on large-scale semantic segmentation datasets, which in…

Material ClassificationObjectSegmentationSemantic Segmentation

Reducing Texture Bias of Deep Neural Networks via Edge Enhancing Diffusion

2024-02-14 · Edgar Heinert, Matthias Rottmann, Kira Maag, Karsten Kahl

Convolutional neural networks (CNNs) for image processing tend to focus on localized texture patterns, commonly referred to as texture bias. While most of the previous works in the literature focus on the task of image c…

Adversarial RobustnessDomain Generalizationimage-classificationImage Classification+2