MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Designed lightweight mobile models using depthwise separable convolutions with two simple global hyperparameters: width multiplier and resolution multiplier.

Serves as qualified reference for deploying task-image-classification in production.

Assumptions

• Spatial feature coherence and data manifold structure adhere to continuous representation hypotheses

Limitations

• Computational complexity scales with spatial resolution and parameter capacity