Equivariant Imaging: Learning Beyond the Range Space

Dongdong Chen, Julian Tachella, Mike E. Davies; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 4379-4388


In various imaging problems, we only have access to compressed measurements of the underlying signals, hindering most learning-based strategies which usually require pairs of signals and associated measurements for training. Learning only from compressed measurements is impossible in general, as the compressed observations do not contain information outside the range of the forward sensing operator. We propose a new end-to-end self-supervised framework that overcomes this limitation by exploiting the equivariances present in natural signals. Our proposed learning strategy performs as well as fully supervised methods. Experiments demonstrate the potential of this framework on inverse problems including sparse-view X-ray computed tomography on real clinical data and image inpainting on natural images. Code has been made available at: https://github.com/edongdongchen/EI.

Related Material

[pdf] [supp] [arXiv]
@InProceedings{Chen_2021_ICCV, author = {Chen, Dongdong and Tachella, Julian and Davies, Mike E.}, title = {Equivariant Imaging: Learning Beyond the Range Space}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2021}, pages = {4379-4388} }