MixConv: Mixed Depthwise Convolutional Kernels

Mingxing TanQuoc V. Le

   Papers with code   Abstract  PDF

Depthwise convolution is becoming increasingly popular in modern efficient ConvNets, but its kernel size is often overlooked. In this paper, we systematically study the impact of different kernel sizes, and observe that combining the benefits of multiple kernel sizes can lead to better accuracy and efficiency... (read more)

Benchmarked Models