Lightweight Unpaired Smartphone ISP Transfer with Semantic Pseudo-Pairing

TL;DR

This paper introduces a lightweight, semantic pseudo-pairing approach for unpaired smartphone ISP transfer, leveraging global context reconstruction and semantic matching to improve color rendering without requiring paired data.

Contribution

The method combines global context reconstruction, semantic embeddings, and optimal transport to generate pseudo pairs, enabling stable training of a compact CNN for unpaired smartphone ISP.

Findings

Achieved 22.569 PSNR and 0.675 SSIM on challenge test set.

Significantly outperformed baseline methods.

Secured 3rd place in SSIM and ΔE among challenge entries.

Abstract

Unpaired smartphone ISP is a challenging problem due to the lack of scene and color alignment between RAW and target RGB images. Many existing methods either require paired data or rely heavily on adversarial training, which can become unstable in the unpaired setting. In this work, we present a simple and effective approach developed for the NTIRE 2026 Learned Smartphone ISP Challenge with Unpaired Data. Our method first reconstructs larger images from training patches to recover global context. Then, we extract semantic embeddings with DINOv2, and use fused Gromov-Wasserstein (FGW) optimal transport to build pseudo pairs between RAW and RGB images at both image and patch levels. This semantic matching allows us to partially alleviate the unpairedness of the data and build these pseudo input-target pairs. Based on these pseudo pairs, we train a lightweight CNN with only 7K parameters for color rendering. The network is designed to be compact and focus on color transformation rather than structural change, which helps reduce artifacts and improve training stability. Our challenge submission achieves 22.569 PSNR, 0.675 SSIM, and 8.067 $\Delta E$ on the final hidden test set, significantly improving over the baseline and achieving the 3rd best SSIM and $\Delta E$ among all challenge entries. Our code is available at github.com/nuniniyujin/Unpaired-ISP .