# Hybrid Loss for Learning Single-Image-based HDR Reconstruction

arXiv: https://arxiv.org/abs/1812.07134 · HTML: https://mlsanity.com/paper/1812.07134/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Dec 2018.

## Built on by (most cited first, 7 of 7)

- [HDR Video Reconstruction: A Coarse-to-fine Network and A Real-world Benchmark Dataset](https://mlsanity.com/paper/2103.14943/) (2021)
- [Deep Learning for HDR Imaging: State-of-the-Art and Future Trends](https://mlsanity.com/paper/2110.10394/) (2021)
- [Single-Image HDR Reconstruction by Multi-Exposure Generation](https://mlsanity.com/paper/2210.15897/) (2022)
- [HDR-cGAN: Single LDR to HDR Image Translation using Conditional GAN](https://mlsanity.com/paper/2110.01660/) (2021)
- [LHDR: HDR Reconstruction for Legacy Content using a Lightweight DNN](https://mlsanity.com/paper/2211.11270/) (2022)
- [Deep chroma compression of tone-mapped images](https://mlsanity.com/paper/2409.16032/) (2024)
- [Image color correction, enhancement, and editing](https://mlsanity.com/paper/2107.13117/) (2021)

## Abstract (arXiv)

This paper tackles high-dynamic-range (HDR) image reconstruction given only a single low-dynamic-range (LDR) image as input. While the existing methods focus on minimizing the mean-squared-error (MSE) between the target and reconstructed images, we minimize a hybrid loss that consists of perceptual and adversarial losses in addition to HDR-reconstruction loss. The reconstruction loss instead of MSE is more suitable for HDR since it puts more weight on both over- and under- exposed areas. It makes the reconstruction faithful to the input. Perceptual loss enables the networks to utilize knowledge about objects and image structure for recovering the intensity gradients of saturated and grossly quantized areas. Adversarial loss helps to select the most plausible appearance from multiple solutions. The hybrid loss that combines all the three losses is calculated in logarithmic space of image intensity so that the outputs retain a large dynamic range and meanwhile the learning becomes tractable. Comparative experiments conducted with other state-of-the-art methods demonstrated that our method produces a leap in image quality.