ChannelNets: Compact and Efficient Convolutional Neural Networks via Channel-Wise Convolutions
This work addresses model efficiency for mobile and embedded AI applications, representing an incremental improvement with a novel focus on compressing the classification layer.
The authors tackled the problem of large model sizes in CNNs for resource-limited applications by proposing ChannelNets, which use channel-wise convolutions to compress models, achieving significant reductions in parameters and computational cost without accuracy loss on ImageNet.
Convolutional neural networks (CNNs) have shown great capability of solving various artificial intelligence tasks. However, the increasing model size has raised challenges in employing them in resource-limited applications. In this work, we propose to compress deep models by using channel-wise convolutions, which re- place dense connections among feature maps with sparse ones in CNNs. Based on this novel operation, we build light-weight CNNs known as ChannelNets. Channel- Nets use three instances of channel-wise convolutions; namely group channel-wise convolutions, depth-wise separable channel-wise convolutions, and the convolu- tional classification layer. Compared to prior CNNs designed for mobile devices, ChannelNets achieve a significant reduction in terms of the number of parameters and computational cost without loss in accuracy. Notably, our work represents the first attempt to compress the fully-connected classification layer, which usually accounts for about 25% of total parameters in compact CNNs. Experimental results on the ImageNet dataset demonstrate that ChannelNets achieve consistently better performance compared to prior methods.