An Enhanced Hybrid MobileNet
This work addresses efficiency improvements for image recognition on mobile devices, representing an incremental enhancement to existing MobileNet methods.
The authors tackled the problem of reducing computational costs for neural networks on mobile devices by proposing an enhanced MobileNet architecture that replaces the resolution multiplier with a depth multiplier combined with pooling methods, achieving simultaneous reduction in computational cost and increased accuracy on CIFAR datasets.
Complicated and deep neural network models can achieve high accuracy for image recognition. However, they require a huge amount of computations and model parameters, which are not suitable for mobile and embedded devices. Therefore, MobileNet was proposed, which can reduce the number of parameters and computational cost dramatically. The main idea of MobileNet is to use a depthwise separable convolution. Two hyper-parameters, a width multiplier and a resolution multiplier are used to the trade-off between the accuracy and the latency. In this paper, we propose a new architecture to improve the MobileNet. Instead of using the resolution multiplier, we use a depth multiplier and combine with either Fractional Max Pooling or the max pooling. Experimental results on CIFAR database show that the proposed architecture can reduce the amount of computational cost and increase the accuracy simultaneously.