# Raising The Limit Of Image Rescaling Using Auxiliary Encoding

arXiv: https://arxiv.org/abs/2303.06747 · HTML: https://mlsanity.com/paper/2303.06747/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 20 references that are in the map.
- First posted on arXiv 12 Mar 2023.

## Builds on (most cited first, 10 of 10)

- [Image Super-Resolution Using Very Deep Residual Channel Attention Networks](https://mlsanity.com/paper/1807.02758/) (2018)
- [Blind Super-Resolution Kernel Estimation using an Internal-GAN](https://mlsanity.com/paper/1909.06581/) (2019)
- [Invertible Image Rescaling](https://mlsanity.com/paper/2005.05650/) (2020)
- [Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling](https://mlsanity.com/paper/2108.05301/) (2021)
- [Learned Image Downscaling for Upscaling using Content Adaptive Resampler](https://mlsanity.com/paper/1907.12904/) (2019)
- [Self-Conditioned Probabilistic Learning of Video Rescaling](https://mlsanity.com/paper/2107.11639/) (2021)
- [Towards Bidirectional Arbitrary Image Rescaling: Joint Optimization and Cycle Idempotence](https://mlsanity.com/paper/2203.00911/) (2022)
- [Scale-arbitrary Invertible Image Downscaling](https://mlsanity.com/paper/2201.12576/) (2022)
- [Approaching the Limit of Image Rescaling via Flow Guidance](https://mlsanity.com/paper/2111.05133/) (2021)
- [Enhancing Image Rescaling using Dual Latent Variables in Invertible Neural Network](https://mlsanity.com/paper/2207.11844/) (2022)

## Abstract (arXiv)

Normalizing flow models using invertible neural networks (INN) have been widely investigated for successful generative image super-resolution (SR) by learning the transformation between the normal distribution of latent variable z and the conditional distribution of high-resolution (HR) images gave a low-resolution (LR) input. Recently, image rescaling models like IRN utilize the bidirectional nature of INN to push the performance limit of image upscaling by optimizing the downscaling and upscaling steps jointly. While the random sampling of latent variable z is useful in generating diverse photo-realistic images, it is not desirable for image rescaling when accurate restoration of the HR image is more important. Hence, in places of random sampling of z, we propose auxiliary encoding modules to further push the limit of image rescaling performance. Two options to store the encoded latent variables in downscaled LR images, both readily supported in existing image file format, are proposed. One is saved as the alpha-channel, the other is saved as meta-data in the image header, and the corresponding modules are denoted as suffixes -A and -M respectively. Optimal network architectural changes are investigated for both options to demonstrate their effectiveness in raising the rescaling performance limit on different baseline models including IRN and DLV-IRN.