arXiv · 2611.13504
Dark Matters: VAE-Aware HDR Encoding for SDR-to-HDR Video Reconstruction
Abstract
SDR-to-HDR reconstruction is an ill-posed problem in which clipped highlights, crushed shadows, and quantization artifacts must be recovered while preserving the original content. Latent diffusion models provide powerful generative priors for restoring missing detail, but their effectiveness is constrained by how HDR signals pass through the pretrained VAE. We find that common log-like encodings suppress low-luminance chromatic differences before generation, producing a dark-side collapse in the frozen-VAE latent. We introduce EquivFlow, a compact, invertible exposure-equivariant encoding that maps exposure changes to additive shifts while retaining a scale-independent chromatic component, without modifying the VAE. Encoding-level tests across pretrained image and video VAEs show that its exposure structure and deep-dark chroma transport persist across the tested hosts. Integrated into SDR-to-HDR video diffusion, EquivFlow delivers consistent chromatic-fidelity gains across the tested settings while retaining competitive reconstruction fidelity.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Donghoon Cha, Bomseok Park, Hansl von Kwon, Ku DoHyung, Hanwool Sul. 2026-11-10. Dark Matters: VAE-Aware HDR Encoding for SDR-to-HDR Video Reconstruction. https://searcharxiv.com/abs/2611.13504
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.