2019/05/12 by Nichita Diaconu, Diaconu, Nichita, Daniel E. Worrall +2 · 2 citations
Computer Science · Mathematics · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Multimodal Machine Learning Applications #cs.CV #cs.LG #stat.ML
paper · pdf · doi:10.48550/arxiv.1905.04663
Accepted to ICML 2019
arxiv created 2019/05/12 · openalex publication_date 2019/05/12 · arxiv updated 2019/05/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Recent work (Cohen & Welling, 2016) has shown that generalizations of convolutions, based on group theory, provide powerful inductive biases for learning. In these generalizations, filters are not only translated but can also be rotated, flipped, etc. However, coming up with exact models of how to rotate a 3 x 3 filter on a square pixel-grid is difficult. In this paper, we learn how to transform filters for use in the group convolution, focussing on roto-translation. For this, we learn a filter basis and all rotated versions of that filter basis. Filters are then encoded by a set of rotation invariant coefficients. To rotate a filter, we switch the basis. We demonstrate we can produce feature maps with low sensitivity to input rotations, while achieving high performance on MNIST and CIFAR-10.