ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias

Open paper
Benchmark Model Rank Results
video-object-segmentation-on-davis-2017ViTAE-T-Stage#1Jaccard (Mean): 79.4J&F: 82.5F-Score: 85.5
video-object-segmentation-on-davis-2016ViTAE-T-Stage#3J&F: 89.8F-Score: 90.4Jaccard (Mean): 89.2
image-classification-on-flowers-102ViTAE-S#42Accuracy: 97.8
image-classification-on-cifar-10ViTAE-T#95Percentage correct: 97.7
image-classification-on-cifar-100ViTAE-T#96Percentage correct: 85.0
semantic-segmentation-on-ade20kViTAE-S-Stage + UPerNet#173Validation mIoU: 47.8
image-classification-on-imagenetViTAE-B-Stage#443Top 1 Accuracy: 83.6%Number of params: 48.5MGFLOPs: 27.6
image-classification-on-imagenetViTAE-S-Stage#603Top 1 Accuracy: 82.2%Number of params: 19.2MGFLOPs: 12.0
image-classification-on-imagenetViTAE-13M#711Top 1 Accuracy: 81%Number of params: 13.2MGFLOPs: 6.8
image-classification-on-imagenetViTAE-6M#896Top 1 Accuracy: 77.9%Number of params: 6.5MGFLOPs: 4
image-classification-on-imagenetViTAE-T-Stage#939Top 1 Accuracy: 76.8%Number of params: 4.8MGFLOPs: 4.6
image-classification-on-imagenetViTAE-T#995Top 1 Accuracy: 75.3%GFLOPs: 3.0

Rank counts only results with a code link.