X2HDR: HDR Image Generation in a Perceptually Uniform Space

TL;DR

This paper introduces X2HDR, a method that adapts pretrained diffusion models for HDR image generation by converting HDR inputs into perceptually uniform spaces, enhancing fidelity and dynamic range without retraining from scratch.

Contribution

It proposes an efficient adaptation strategy using perceptually uniform encodings to enable HDR generation with existing diffusion models, improving quality and versatility.

Findings

Perceptually uniform encoding improves HDR reconstruction fidelity.

Fine-tuning only the denoiser yields better perceptual results.

The method supports both text-to-HDR synthesis and RAW-to-HDR reconstruction.

Abstract

High-dynamic-range (HDR) formats and displays are becoming increasingly prevalent, yet state-of-the-art image generators (e.g., Stable Diffusion and FLUX) typically remain limited to low-dynamic-range (LDR) output due to the lack of large-scale HDR training data. In this work, we show that existing pretrained diffusion models can be easily adapted to HDR generation without retraining from scratch. A key challenge is that HDR images are natively represented in linear RGB, whose intensity and color statistics differ substantially from those of sRGB-encoded LDR images. This gap, however, can be effectively bridged by converting HDR inputs into perceptually uniform encodings (e.g., using PU21 or PQ). Empirically, we find that LDR-pretrained variational autoencoders (VAEs) reconstruct PU21-encoded HDR inputs with fidelity comparable to LDR data, whereas linear RGB inputs cause severe degradations. Motivated by this finding, we describe an efficient adaptation strategy that freezes the VAE and finetunes only the denoiser via low-rank adaptation in a perceptually uniform space. This results in a unified computational method that supports both text-to-HDR synthesis and single-image RAW-to-HDR reconstruction. Experiments demonstrate that our perceptually encoded adaptation consistently improves perceptual fidelity, text-image alignment, and effective dynamic range, relative to previous techniques.