-
[pdf]
[arXiv]
[bibtex]@InProceedings{Zheng_2021_CVPR, author = {Zheng, Chuanxia and Cham, Tat-Jen and Cai, Jianfei}, title = {The Spatially-Correlative Loss for Various Image Translation Tasks}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2021}, pages = {16407-16417} }
The Spatially-Correlative Loss for Various Image Translation Tasks
Abstract
We propose a novel spatially-correlative loss that is simple, efficient, and yet effective for preserving scene structure consistency while supporting large appearance changes during unpaired image-to-image (I2I) translation. Previous methods attempt this by using pixel-level cycle-consistency or feature-level matching losses, but the domain-specific nature of these losses hinder translation across large domain gaps. To address this, we exploit the spatial patterns of self-similarity as a means of defining scene structure. Our spatially-correlative loss is geared towards only capturing spatial relationships within an image rather than domain appearance. We also introduce a new self-supervised learning method to explicitly learn spatially-correlative maps for each specific translation task. We show distinct improvement over baseline models in all three modes of unpaired I2I translation: single-modal, multi-modal, and even single-image translation. This new loss can easily be integrated into existing network architectures and thus allows wide applicability.
Related Material