IVCVMar 28, 2021

Invertible Image Signal Processing

arXiv:2103.15061v2163 citations
AI Analysis

This addresses the need for RAW data access from compressed sRGB images, which is valuable for image editing and computer vision applications, though it appears incremental as it builds on existing ISP concepts.

The paper tackles the problem of converting sRGB images back to RAW format by designing an invertible image signal processing pipeline that enables high-quality sRGB rendering and near-perfect RAW reconstruction, with experiments showing superior quality compared to alternative methods.

Unprocessed RAW data is a highly valuable image format for image editing and computer vision. However, since the file size of RAW data is huge, most users can only get access to processed and compressed sRGB images. To bridge this gap, we design an Invertible Image Signal Processing (InvISP) pipeline, which not only enables rendering visually appealing sRGB images but also allows recovering nearly perfect RAW data. Due to our framework's inherent reversibility, we can reconstruct realistic RAW data instead of synthesizing RAW data from sRGB images without any memory overhead. We also integrate a differentiable JPEG compression simulator that empowers our framework to reconstruct RAW data from JPEG images. Extensive quantitative and qualitative experiments on two DSLR demonstrate that our method obtains much higher quality in both rendered sRGB images and reconstructed RAW data than alternative methods.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes