SdcNet: A Computation-Efficient CNN for Object Recognition

TL;DR

SdcNet is a new, computation-efficient CNN architecture that uses the SdcBlock module to improve feature extraction for object recognition while reducing computational costs.

Contribution

The paper introduces SdcBlock, a novel convolutional module, and SdcNet, a CNN architecture optimized for efficiency and adaptable to various computational constraints.

Findings

Achieved 5.60% error on CIFAR-10 with 55M Flops

Reduced error to 5.24% with 103M Flops

Confirmed high computational efficiency of SdcNet

Abstract

Extracting features from a huge amount of data for object recognition is a challenging task. Convolution neural network can be used to meet the challenge, but it often requires a large number of computation resources. In this paper, a computation-efficient convolutional module, named SdcBlock, is proposed and based on it, the convolution network SdcNet is introduced for object recognition tasks. In the proposed module, optimized successive depthwise convolutions supported by appropriate data management is applied in order to generate vectors containing high density and more varieties of feature information. The hyperparameters can be easily adjusted to suit varieties of tasks under different computation restrictions without significantly jeopardizing the performance. The experiments have shown that SdcNet achieved an error rate of 5.60% in CIFAR-10 with only 55M Flops and also reduced further the error rate to 5.24% using a moderate volume of 103M Flops. The expected computation efficiency of the SdcNet has been confirmed.