RawHDR: High Dynamic Range Image Reconstruction from a Single Raw Image
This addresses the challenge of generating HDR images for photography and imaging applications, but it is incremental as it builds on existing raw data usage with specific enhancements.
The paper tackles the problem of reconstructing high dynamic range (HDR) images from limited 8-bit low dynamic range data by proposing a method that uses raw sensor data instead, aiming to recover scene details in dark and bright areas. The result is a model that learns exposure masks and uses dual intensity and global spatial guidance, validated on a newly collected dataset with empirical evaluations showing superiority.
High dynamic range (HDR) images capture much more intensity levels than standard ones. Current methods predominantly generate HDR images from 8-bit low dynamic range (LDR) sRGB images that have been degraded by the camera processing pipeline. However, it becomes a formidable task to retrieve extremely high dynamic range scenes from such limited bit-depth data. Unlike existing methods, the core idea of this work is to incorporate more informative Raw sensor data to generate HDR images, aiming to recover scene information in hard regions (the darkest and brightest areas of an HDR scene). To this end, we propose a model tailor-made for Raw images, harnessing the unique features of Raw data to facilitate the Raw-to-HDR mapping. Specifically, we learn exposure masks to separate the hard and easy regions of a high dynamic scene. Then, we introduce two important guidances, dual intensity guidance, which guides less informative channels with more informative ones, and global spatial guidance, which extrapolates scene specifics over an extended spatial domain. To verify our Raw-to-HDR approach, we collect a large Raw/HDR paired dataset for both training and testing. Our empirical evaluations validate the superiority of the proposed Raw-to-HDR reconstruction model, as well as our newly captured dataset in the experiments.