MobileOne: An Improved One millisecond Mobile Backbone

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetMobileOne-S4 (distill)#620Top 1 Accuracy: 81.4%GFLOPs: 2.9
image-classification-on-imagenetMobileOne-S4#719Top 1 Accuracy: 79.4%Number of params: 14.8MGFLOPs: 2.978
image-classification-on-imagenetMobileOne-S2 (distill)#737Top 1 Accuracy: 79.1%
image-classification-on-imagenetMobileOne-S3#803Top 1 Accuracy: 78.1%Number of params: 10.1MGFLOPs: 1.896
image-classification-on-imagenetMobileOne-S2#824Top 1 Accuracy: 77.4%Number of params: 7.8MGFLOPs: 1.299
image-classification-on-imagenetMobileOne-S1#871Top 1 Accuracy: 75.9%Number of params: 4.8MGFLOPs: 0.825
image-classification-on-imagenetMobileOne-S0 (distill)#935Top 1 Accuracy: 72.5%Number of params: 2.1MGFLOPs: 0.275
image-classification-on-imagenetMobileOne-S0#947Top 1 Accuracy: 71.4%Number of params: 2.1M

Rank counts only results with a code link.