Grouped Convolution

Introduced by Krizhevsky et al. in ImageNet Classification with Deep Convolutional Neural Networks

A Grouped Convolution uses a group of convolutions - multiple kernels per layer - resulting in multiple channel outputs per layer. This leads to wider networks helping a network learn a varied set of low level and high level features. The original motivation of using Grouped Convolutions in AlexNet was to distribute the model over multiple GPUs as an engineering compromise. But later, with models such as ResNeXt, it was shown this module could be used to improve classification accuracy. Specifically by exposing a new dimension through grouped convolutions, cardinality (the size of set of transformations), we can increase accuracy by increasing it.

Source: ImageNet Classification with Deep Convolutional Neural Networks


Paper Code Results Date Stars


Task Papers Share
General Classification 77 12.07%
Image Classification 71 11.13%
Object Detection 67 10.50%
Quantization 41 6.43%
Semantic Segmentation 36 5.64%
Object Recognition 22 3.45%
Instance Segmentation 17 2.66%
Model Compression 15 2.35%
Network Pruning 9 1.41%