[论文] DiffHDR: Re-Exposing LDR Videos with Video Diffusion Models

## 论文概要 **研究领域**: CV **作者**: Zhengming Yu, Li Ma, Mingming He **发布时间**: 2025-04-08 **arXiv**: [2504.06259](https://arxiv.org/abs/2504.06259) ## 中文摘要 大多数数字视频以8位低动态范围（LDR）格式存储，由于饱和和量化，原始高动态范围（HDR）场景的辐射度大量丢失。这种高光和阴影细节的丢失阻碍了对HDR显示器的精确亮度映射，并限制了后期制作流程中有意义的重新曝光。虽然已有技术提出通过动态范围扩展将LDR图像转换为HDR，但它们在恢复过曝和欠曝区域的真实细节方面存在困难。为此，本文提出DiffHDR，一个将LDR到HDR转换形式化为视频扩散模型潜在空间中的生成式辐射修复任务的框架。通过在Log-Gamma颜色空间中操作，DiffHDR利用预训练视频扩散模型的时空生成先验，在过曝和欠曝区域合成合理的HDR辐射度，同时恢复量化像素的连续场景辐射度。该框架还支持由文本提示或参考图像引导的可控LDR到HDR视频转换。为应对配对HDR视频数据的稀缺，我们开发了从静态HDRI图合成高质量HDR视频训练数据的流水线。大量实验表明，DiffHDR在辐射度保真度和时间稳定性方面显著超越最先进方法，产生具有相当大重新曝光自由度的真实感HDR视频。 ## 原文摘要 Most digital videos are stored in 8-bit low dynamic range (LDR) formats, where much of the original high dynamic range (HDR) scene辐射 is lost due to saturation and quantization. This loss of highlight and shadow detail precludes mapping accurate luminance to HDR displays and limits meaningful re-exposure in post-production workflows. Although techniques have been proposed to convert LDR images to HDR through dynamic range expansion, they struggle to restore realistic detail in the over- and underexposed regions. To address this, we present DiffHDR, a framework that formulates LDR-to-HDR conversion as a generative radiance inpainting task within the latent space of a video diffusion model. --- *自动采集于 2026-04-09* #论文 #arXiv #CV #小凯