ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression

• URL: http://arxiv.org/abs/2509.20234v3
• Date: Tue, 28 Oct 2025 11:26:53 GMT
• Title: ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression
• Abstract summary: We revisit the hypothesis that Convolutional Neural Networks (CNNs) are inherently texture-biased.<n>We propose a domain-agnostic framework that quantifies feature reliance through systematic suppression of shape, texture, and color cues.<n>We extend the analysis across computer vision, medical imaging, and remote sensing, revealing that reliance patterns differ systematically.
• Score: 10.20881592893918
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The hypothesis that Convolutional Neural Networks (CNNs) are inherently texture-biased has shaped much of the discourse on feature use in deep learning. We revisit this hypothesis by examining limitations in the cue-conflict experiment by Geirhos et al. To address these limitations, we propose a domain-agnostic framework that quantifies feature reliance through systematic suppression of shape, texture, and color cues, avoiding the confounds of forced-choice conflicts. By evaluating humans and neural networks under controlled suppression conditions, we find that CNNs are not inherently texture-biased but predominantly rely on local shape features. Nonetheless, this reliance can be substantially mitigated through modern training strategies or architectures (ConvNeXt, ViTs). We further extend the analysis across computer vision, medical imaging, and remote sensing, revealing that reliance patterns differ systematically: computer vision models prioritize shape, medical imaging models emphasize color, and remote sensing models exhibit a stronger reliance on texture. Code is available at https://github.com/tomburgert/feature-reliance.

Related papers

• Perceptual misalignment of texture representations in convolutional neural networks [3.68986335765876]
  We compare the perceptual content captured by feature correlations computed for a diverse pool of convolutional neural networks (CNNs)<n>We find no connection between conventional measures of CNN quality as a model of the visual system and its alignment with human texture perception.<n>We conclude that texture perception involves mechanisms that are distinct from those that are commonly modeled using approaches based on CNNs trained on object recognition.
  arXiv  Detail & Related papers  (2026-04-01T19:51:45Z)
• Concept-Guided Interpretability via Neural Chunking [64.6429903327095]
  We show that neural networks exhibit patterns in their raw population activity that mirror regularities in the training data.<n>We propose three methods to extract recurring chunks on a neural population level.<n>Our work points to a new direction for interpretability, one that harnesses both cognitive principles and the structure of naturalistic data.
  arXiv  Detail & Related papers  (2025-05-16T13:49:43Z)
• Feature CAM: Interpretable AI in Image Classification [2.4409988934338767]
  There is a lack of trust to use Artificial Intelligence in critical and high-precision fields such as security, finance, health, and manufacturing industries. We introduce a novel technique Feature CAM, which falls in the perturbation-activation combination, to create fine-grained, class-discriminative visualizations. The resulting saliency maps proved to be 3-4 times better human interpretable than the state-of-the-art in ABM.
  arXiv  Detail & Related papers  (2024-03-08T20:16:00Z)
• BreakingBED -- Breaking Binary and Efficient Deep Neural Networks by Adversarial Attacks [65.2021953284622]
  We study robustness of CNNs against white-box and black-box adversarial attacks. Results are shown for distilled CNNs, agent-based state-of-the-art pruned models, and binarized neural networks.
  arXiv  Detail & Related papers  (2021-03-14T20:43:19Z)
• Assessing The Importance Of Colours For CNNs In Object Recognition [70.70151719764021]
  Convolutional neural networks (CNNs) have been shown to exhibit conflicting properties. We demonstrate that CNNs often rely heavily on colour information while making a prediction. We evaluate a model trained with congruent images on congruent, greyscale, and incongruent images.
  arXiv  Detail & Related papers  (2020-12-12T22:55:06Z)
• Informative Dropout for Robust Representation Learning: A Shape-bias Perspective [84.30946377024297]
  We propose a light-weight model-agnostic method, namely Informative Dropout (InfoDrop), to improve interpretability and reduce texture bias. Specifically, we discriminate texture from shape based on local self-information in an image, and adopt a Dropout-like algorithm to decorrelate the model output from the local texture.
  arXiv  Detail & Related papers  (2020-08-10T16:52:24Z)
• Neural Networks with Recurrent Generative Feedback [61.90658210112138]
  We instantiate this design on convolutional neural networks (CNNs) In the experiments, CNN-F shows considerably improved adversarial robustness over conventional feedforward CNNs on standard benchmarks.
  arXiv  Detail & Related papers  (2020-07-17T19:32:48Z)
• Teaching CNNs to mimic Human Visual Cognitive Process & regularise Texture-Shape bias [18.003188982585737]
  Recent experiments in computer vision demonstrate texture bias as the primary reason for supreme results in models employing Convolutional Neural Networks (CNNs) It is believed that the cost function forces the CNN to take a greedy approach and develop a proclivity for local information like texture to increase accuracy, thus failing to explore any global statistics. We propose CognitiveCNN, a new intuitive architecture, inspired from feature integration theory in psychology to utilise human interpretable feature like shape, texture, edges etc. to reconstruct, and classify the image.
  arXiv  Detail & Related papers  (2020-06-25T22:32:54Z)
• The shape and simplicity biases of adversarially robust ImageNet-trained CNNs [9.707679445925516]
  We study the shape bias and internal mechanisms that enable the generalizability of AlexNet, GoogLeNet, and ResNet-50 models trained via adversarial training. Remarkably, adversarial training induces three simplicity biases into hidden neurons in the process of "robustifying" CNNs.
  arXiv  Detail & Related papers  (2020-06-16T16:38:16Z)
• Hold me tight! Influence of discriminative features on deep network boundaries [63.627760598441796]
  We propose a new perspective that relates dataset features to the distance of samples to the decision boundary. This enables us to carefully tweak the position of the training samples and measure the induced changes on the boundaries of CNNs trained on large-scale vision datasets.
  arXiv  Detail & Related papers  (2020-02-15T09:29:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.