SO(2) and O(2) Equivariance in Image Recognition with Bessel-Convolutional Neural Networks
This work addresses the need for more robust image analysis models that can handle arbitrary orientations and reflections, which is important for computer vision applications, but it is incremental as it builds on existing B-CNN frameworks.
The paper tackles the problem of incorporating rotation and reflection equivariances in image recognition by extending Bessel-convolutional neural networks (B-CNNs) to include these symmetries, resulting in improved performance compared to other methods as shown in extensive studies.
For many years, it has been shown how much exploiting equivariances can be beneficial when solving image analysis tasks. For example, the superiority of convolutional neural networks (CNNs) compared to dense networks mainly comes from an elegant exploitation of the translation equivariance. Patterns can appear at arbitrary positions and convolutions take this into account to achieve translation invariant operations through weight sharing. Nevertheless, images often involve other symmetries that can also be exploited. It is the case of rotations and reflections that have drawn particular attention and led to the development of multiple equivariant CNN architectures. Among all these methods, Bessel-convolutional neural networks (B-CNNs) exploit a particular decomposition based on Bessel functions to modify the key operation between images and filters and make it by design equivariant to all the continuous set of planar rotations. In this work, the mathematical developments of B-CNNs are presented along with several improvements, including the incorporation of reflection and multi-scale equivariances. Extensive study is carried out to assess the performances of B-CNNs compared to other methods. Finally, we emphasize the theoretical advantages of B-CNNs by giving more insights and in-depth mathematical details.