Search papers, labs, and topics across Lattice.
This paper introduces a novel image-to-image translation (I2I) framework that incorporates rotation symmetry priors using rotation group equivariant convolutions. They further propose Transformation Learnable Convolutions (TL-Conv) to adaptively learn transformation groups, enhancing symmetry preservation across diverse datasets. Theoretical analysis demonstrates TL-Conv maintains exact equivariance in continuous domains and provides error bounds for discrete cases, validated by superior performance across various I2I tasks.
Achieve state-of-the-art image-to-image translation by baking in rotation symmetry priors, a surprisingly effective inductive bias.
Image-to-image translation (I2I) is a fundamental task in computer vision, focused on mapping an input image from a source domain to a corresponding image in a target domain while preserving domain-invariant features and adapting domain-specific attributes. Despite the remarkable success of deep learning-based I2I approaches, the lack of paired data and unsupervised learning framework still hinder their effectiveness. In this work, we address the challenge by incorporating transformation symmetry priors into image-to-image translation networks. Specifically, we introduce rotation group equivariant convolutions to achieve rotation equivariant I2I framework, a novel contribution, to the best of our knowledge, along this research direction. This design ensures the preservation of rotation symmetry, one of the most intrinsic and domain-invariant properties of natural and scientific images, throughout the network. Furthermore, we conduct a systematic study on image symmetry priors on real dataset and propose a novel transformation learnable equivariant convolutions (TL-Conv) that adaptively learns transformation groups, enhancing symmetry preservation across diverse datasets. We also provide a theoretical analysis of the equivariance error of TL-Conv, proving that it maintains exact equivariance in continuous domains and provide a bound for the error in discrete cases. Through extensive experiments across a range of I2I tasks, we validate the effectiveness and superior performance of our approach, highlighting the potential of equivariant networks in enhancing generation quality and its broad applicability. Our code is available at https://github.com/tanfy929/Equivariant-I2I
