Dilated convolution with learnable spacings

Abstract

Recent works indicate that convolutional neural networks (CNN) need large receptive fields (RF) to compete with visual transformers and their attention mechanism. In CNNs, RFs can simply be enlarged by increasing the convolution kernel sizes. Yet the number of trainable parameters, which scales quadratically with the kernel's size in the 2D case, rapidly becomes prohibitive, and the training is notoriously difficult. This paper presents a new method to increase the RF size without increasing the number of parameters. The dilated convolution (DC) has already been proposed for the same purpose. DC can be seen as a convolution with a kernel that contains only a few non-zero elements placed on a regular grid. Here we present a new version of the DC in which the spacings between the non-zero elements, or equivalently their positions, are no longer fixed but learnable via backpropagation thanks to an interpolation technique. We call this method "Dilated Convolution with Learnable Spacings" (DCLS) and generalize it to the n-dimensional convolution case. However, our main focus here will be on the 2D case. We first tried our approach on ResNet50: we drop-in replaced the standard convolutions with DCLS ones, which increased the accuracy of ImageNet1k classification at iso-parameters, but at the expense of the throughput. Next, we used the recent ConvNeXt state-of-the-art convolutional architecture and drop-in replaced the depthwise convolutions with DCLS ones. This not only increased the accuracy of ImageNet1k classification but also of typical downstream and robustness tasks, again at iso-parameters but this time with negligible cost on throughput, as ConvNeXt uses separable convolutions. Conversely, classic DC led to poor performance with both ResNet50 and ConvNeXt. The code of the method is available at: https://github.com/K-H-Ismail/Dilated-Convolution-with-Learnable-Spacings-PyTorch.

Published in The Eleventh International Conference on Learning Representations (ICLR) 2023. (https://openreview.net/forum?id=Q3-1vRh3HOA)

References in corpus (9)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• WaveNet: A Generative Model for Raw Audio
• BEiT: BERT Pre-Training of Image Transformers
• More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsity
• Deep Scale-spaces: Equivariance Over Scale
• Scale-Equivariant Steerable Networks
• FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes
• A Pre-defined Sparse Kernel Based Convolution for Deep CNNs
• DISCO: accurate Discrete Scale Convolutions

Cited by in corpus (3)

• LDConv: Linear deformable convolution for improving convolutional neural networks
• Optical flow estimation from event-based cameras and spiking neural networks
• CB-HVTNet: A channel-boosted hybrid vision transformer network for lymphocyte assessment in histopathological images