Stylized ImageNet
Introduced by Geirhos et al. in ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
The Stylized-ImageNet dataset is created by removing local texture cues in ImageNet while retaining global shape information on natural images via AdaIN style transfer. This nudges CNNs towards learning more about shapes and less about local textures.
Source: Adversarial Examples Improve Image RecognitionBenchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
Stylized ImageNet
|
DeiT-S
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
