LatentHDR: Decoupling Exposure from Diffusion via Conditional Latent-to-Latent Mapping for Text/Image-to-Panoramic HDR

TL;DR

LatentHDR introduces a novel latent space approach for efficient, structurally consistent HDR image synthesis from text or images, outperforming diffusion-based methods in quality and computational cost.

Contribution

It decouples scene and exposure modeling in latent space using a pretrained diffusion backbone and a lightweight mapping head, enabling scalable HDR generation.

Findings

Achieves state-of-the-art dynamic range on benchmarks.

Reduces computational cost by an order of magnitude.

Produces structurally consistent exposure stacks in a single pass.

Abstract

High Dynamic Range (HDR) generation remains challenging for generative models, which are largely limited to low dynamic range outputs. Recent diffusionbased approaches approximate HDR by generating multiple exposure-conditioned samples, incurring high computational cost and structural inconsistencies across exposures. We propose LatentHDR, a framework that decouples scene generation from exposure modeling in latent space. A pretrained diffusion backbone produces a single coherent scene representation, while a lightweight conditional latent to-latent head deterministically maps it to exposure-specific representations. This enables the generation of a dense, structurally consistent exposure stack in a single pass. This design eliminates multi-pass diffusion, ensures cross-exposure alignment, and enables scalable HDR synthesis. LatentHDR supports both textand image-conditioned HDR generation for perspective and panoramic scenes. Experiments on synthetic data and the SI-HDR benchmark show that LatentHDR achieves state-of-the-art dynamic range with competitive perceptual quality, while reducing computation by an order of magnitude. Our results demonstrate that high-quality HDR generation can be achieved through structured latent modeling, challenging the need for stochastic multi-exposure generation.