“Scaling Up Your Kernels to 31×31: Revisiting Large Kernel Design in CNNs (RepLKNet)”, 2022-03-13 (; backlinks; similar):
We revisit large kernel design in modern convolutional neural networks (CNNs).
Inspired by recent advances in vision transformers (ViTs), in this paper, we demonstrate that using a few large convolutional kernels instead of a stack of small kernels could be a more powerful paradigm. We suggested 5 guidelines, eg. applying re-parameterized large depth-wise convolutions, to design efficient high-performance large-kernel CNNs.
Following the guidelines, we propose RepLKNet, a pure CNN architecture whose kernel size is as large as 31×31, in contrast to commonly used 3×3.
RepLKNet greatly closes the performance gap between CNNs and ViTs, eg. achieving comparable or superior results than Swin Transformer on ImageNet and a few typical downstream tasks, with lower latency. RepLKNet also shows nice scalability to big data and large models, obtaining 87.8% top-1 accuracy on ImageNet and 56.0% mIoU on ADE20K, which is very competitive among the state-of-the-arts with similar model sizes. Our study further reveals that, in contrast to small-kernel CNNs, large-kernel CNNs have much larger effective receptive fields and higher shape bias rather than texture bias.
Code & models at Github.