Tackling Ill-Posedness of Reversible Image Conversion With Well-Posed Invertible Network
- 29 January 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (6) , 6847-6864
Abstract
Reversible image conversion (RIC) suffers from ill-posedness issues due to its forward conversion process being considered an underdetermined system. Despite employing invertible neural networks (INN), existing RIC methods intrinsically remain ill-posed as inevitably introducing uncertainty by incorporating randomly sampled variables. To tackle the ill-posedness dilemma, we focus on developing a reliable approximate left inverse for the underdetermined system by constructing an overdetermined system with a non-zero Gram determinant, thus ensuring a well-posed solution. Based on this principle, we propose a well-posed invertible $1\times 1$ convolution (WIC), which eliminates the reliance on random variable sampling and enables the development of well-posed invertible networks. Furthermore, we design two innovative networks, WIN-Naïve and WIN, with the latter incorporating advanced skip-connections to enhance long-term memory. Our methods are evaluated across diverse RIC tasks, including reversible image hiding, image rescaling, and image decolorization, consistently achieving state-of-the-art performance. Extensive experiments validate the effectiveness of our approach, demonstrating its ability to overcome the bottlenecks of existing RIC solutions and setting a new benchmark in the field.
Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (62131003)
- National Natural Science Foundation of China (U25B2065)
This publication has 69 references indexed in Scilit:
- CD-iNet: Deep Invertible Network for Perceptual Image Color Difference MeasurementInternational Journal of Computer Vision, 2024
- Learning spatiotemporal embedding with gated convolutional recurrent networks for translation initiation site predictionPattern Recognition, 2022
- Invertible Rescaling Network and Its ExtensionsInternational Journal of Computer Vision, 2022
- Task-Aware Image Downscaling2018
- Inverse problems with Poisson data: statistical regularization theory, applications and algorithmsInverse Problems, 2016
- Deep Residual Learning for Image Recognition2016
- Single image super-resolution from transformed self-exemplars2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Microsoft COCO: Common Objects in Context2014
- On Single Image Scale-Up Using Sparse-Representations2012