CoCAViT: Compact Vision Transformer with Robust Global Coordination
Open paper
Benchmark
Model
Rank
Results
object-detection-on-coco-val2017
CoCAViT-21M (Cascade Mask R-CNN 3x)
–
box AP: 51.8
AP50: 70.5
AP75: 56.1
object-detection-on-coco-val2017
CoCAViT-28M (Cascade Mask R-CNN 3x)
–
box AP: 52.2
AP50: 71.0
AP75: 56.8
Rank counts only results with a code link.