MobileOne: An Improved One millisecond Mobile Backbone
This addresses the need for faster and more accurate mobile AI applications, though it is incremental as it builds on prior efficient architectures.
The paper tackles the problem of designing efficient neural network backbones for mobile devices by identifying bottlenecks in existing models and proposing MobileOne, which achieves under 1 ms inference time on an iPhone12 with 75.9% top-1 accuracy on ImageNet, outperforming models like MobileFormer by 38x in speed and EfficientNet by 2.3% in accuracy.
Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9% top-1 accuracy on ImageNet. We show that MobileOne achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38x faster. Our model obtains 2.3% better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device. Code and models are available at https://github.com/apple/ml-mobileone