Dilated Convolution with Learnable Spacings: beyond bilinear interpolation

2023/06/01 by Ismail Khalfaoui-Hassani, Khalfaoui-Hassani, Ismail, Thomas Pellegrini +3 · 2 citations
Computer Science · #Advanced Neural Network Applications #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.2306.00817

openalex publication_date 2023/06/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01

Abstract

Dilated Convolution with Learnable Spacings (DCLS) is a recently proposed variation of the dilated convolution in which the spacings between the non-zero elements in the kernel, or equivalently their positions, are learnable. Non-integer positions are handled via interpolation. Thanks to this trick, positions have well-defined gradients. The original DCLS used bilinear interpolation, and thus only considered the four nearest pixels. Yet here we show that longer range interpolations, and in particular a Gaussian interpolation, allow improving performance on ImageNet1k classification on two state-of-the-art convolutional architectures (ConvNeXt and Conv\-Former), without increasing the number of parameters. The method code is based on PyTorch and is available at https://github.com/K-H-Ismail/Dilated-Convolution-with-Learnable-Spacings-PyTorch

Citations

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Gaussian Mixture Convolution Networks
• Understanding Gaussian Attention Bias of Vision Transformers Using Effective Receptive Fields
• MetaFormer Baselines for Vision
• Dilated Convolution with Learnable Spacings

Cited by

• DelRec: learning delays in recurrent spiking neural networks
• mGRADE: Minimal Recurrent Gating Meets Delay Convolutions for Lightweight Sequence Modeling

Related

• Dilated Convolution with Learnable Spacings
• Audio classification with Dilated Convolution with Learnable Spacings
• Dilated Convolution with Learnable Spacings makes visual models more aligned with humans: a Grad-CAM study
• Generalized Dilation Neural Networks
• Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation