https://github.com/rgeirhos/texture-vs-shape
Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)
https://github.com/rgeirhos/texture-vs-shape
deep-learning features human-vision object-recognition psychophysics shape texture
Last synced: 2 months ago
JSON representation
Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)
- Host: GitHub
- URL: https://github.com/rgeirhos/texture-vs-shape
- Owner: rgeirhos
- License: other
- Created: 2018-11-28T09:56:15.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2023-05-02T13:43:26.000Z (about 2 years ago)
- Last Synced: 2025-03-28T10:10:07.267Z (3 months ago)
- Topics: deep-learning, features, human-vision, object-recognition, psychophysics, shape, texture
- Language: R
- Homepage: https://openreview.net/forum?id=Bygh9j09KX
- Size: 145 MB
- Stars: 794
- Watchers: 18
- Forks: 104
- Open Issues: 0
-
Metadata Files:
- Readme: README.md