RAW Image Reconstruction from RGB on Smartphones. NTIRE 2025 Challenge Report
This work provides a solution for generating realistic RAW data from abundant sRGB images, which is important for low-level vision tasks in computational photography, though it appears incremental as part of an ongoing challenge series.
The paper tackles the problem of reconstructing RAW sensor images from sRGB images on smartphones without metadata, addressing the scarcity of RAW datasets. It reports on the NTIRE 2025 challenge where over 150 participants submitted models, establishing a new state-of-the-art benchmark for this task.
Numerous low-level vision tasks operate in the RAW domain due to its linear properties, bit depth, and sensor designs. Despite this, RAW image datasets are scarce and more expensive to collect than the already large and public sRGB datasets. For this reason, many approaches try to generate realistic RAW images using sensor information and sRGB images. This paper covers the second challenge on RAW Reconstruction from sRGB (Reverse ISP). We aim to recover RAW sensor images from smartphones given the corresponding sRGB images without metadata and, by doing this, ``reverse" the ISP transformation. Over 150 participants joined this NTIRE 2025 challenge and submitted efficient models. The proposed methods and benchmark establish the state-of-the-art for generating realistic RAW data.