Scaling Up Your Kernels to 31×31: Revisiting Large Kernel Design in CNNs

🤖 Plain-English Summary

We revisit large kernel design in modern convolutional neural networks (CNNs). 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.

🔑 Key Findings

• 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 five guidelines, e.g., applying re-parameterized large depthwise 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.

🔗 Related Resources