EdgeNeXt: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applications

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetEdgeNeXt-S#720Top 1 Accuracy: 79.4%Number of params: 5.6MGFLOPs: 2.6
image-classification-on-imagenetEdgeNeXt-XXS#949Top 1 Accuracy: 71.2%Number of params: 1.3MGFLOPs: 0.522
semantic-segmentation-on-pascal-voc-2012EdgeNeXt#28Mean IoU: 80.2%FLOPS: 8.7GParams: 6.5M

Rank counts only results with a code link.