Edge-RecViT: Efficient Vision Transformer via Semantic-Refined Dynamic Recursion

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10Edge-RecViT (Base)#13Percentage correct: 99.16
image-classification-on-cifar-100Edge-RecViT (Base)#35Percentage correct: 90.78
instance-segmentation-on-coco-minivalEdge-RecViT (Base)#43mask AP: 48.3
image-classification-on-imagenetEdge-RecViT (Base)#625Top 1 Accuracy: 82.0%

Rank counts only results with a code link.