Quantized Convolutional Neural Networks for Mobile Devices

TL;DR

This paper introduces Quantized CNNs that significantly speed up computation and reduce memory usage, enabling accurate image classification on mobile devices within one second.

Contribution

It presents a novel quantization framework for CNNs that achieves substantial speed-up and compression with minimal accuracy loss.

Findings

4-6x speed-up on ILSVRC-12 benchmark

15-20x model compression

Only 1% accuracy loss

Abstract

Recently, convolutional neural networks (CNN) have demonstrated impressive performance in various computer vision tasks. However, high performance hardware is typically indispensable for the application of CNN models due to the high computation complexity, which prohibits their further extensions. In this paper, we propose an efficient framework, namely Quantized CNN, to simultaneously speed-up the computation and reduce the storage and memory overhead of CNN models. Both filter kernels in convolutional layers and weighting matrices in fully-connected layers are quantized, aiming at minimizing the estimation error of each layer's response. Extensive experiments on the ILSVRC-12 benchmark demonstrate 4~6x speed-up and 15~20x compression with merely one percentage loss of classification accuracy. With our quantized CNN model, even mobile devices can accurately classify images within one second.