HyperAIHyperAI
HomeNewsLatest PapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAIHyperAI

Object Recognition On Shape Bias

Metrics

shape bias

Results

Performance results of various models on this benchmark

Model Name
shape bias
Paper TitleRepository
ViT-22B-38486.4Scaling Vision Transformers to 22 Billion Parameters-
ResNet-5022.1ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness-
ViT-22B-56083.8Scaling Vision Transformers to 22 Billion Parameters-
GoogLeNet31.2ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness-
ViT-22B-22478.0Scaling Vision Transformers to 22 Billion Parameters-
CLIP (ViT-B)79.9Learning Transferable Visual Models From Natural Language Supervision-
SWSL (ResNet-50)28.6Billion-scale semi-supervised learning for image classification-
SimCLR (ResNet-50x4)40.7A Simple Framework for Contrastive Learning of Visual Representations-
Parti91.7Intriguing properties of generative classifiers-
VGG-1617.2ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness-
Stable Diffusion92.7Intriguing properties of generative classifiers-
SimCLR (ResNet-50x1)38.9A Simple Framework for Contrastive Learning of Visual Representations-
SWSL (ResNeXt-101)49.8Billion-scale semi-supervised learning for image classification-
ResNet-50 (with strong augmentations)62.2The Origins and Prevalence of Texture Bias in Convolutional Neural Networks-
ResNet-50 (L2 eps 5.0 adv trained)69.5Do Adversarially Robust ImageNet Models Transfer Better?-
SimCLR (ResNet-50x2)41.7A Simple Framework for Contrastive Learning of Visual Representations-
Imagen98.7Intriguing properties of generative classifiers-
AlexNet42.9ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness-
0 of 18 row(s) selected.
Object Recognition On Shape Bias | SOTA | HyperAI