Few-Shot Domain Adaptation for Low Light RAW Image Enhancement

27 Mar 2023  ·  K. Ram Prabhakar, Vishal Vinod, Nihar Ranjan Sahoo, R. Venkatesh Babu ·

Enhancing practical low light raw images is a difficult task due to severe noise and color distortions from short exposure time and limited illumination. Despite the success of existing Convolutional Neural Network (CNN) based methods, their performance is not adaptable to different camera domains. In addition, such methods also require large datasets with short-exposure and corresponding long-exposure ground truth raw images for each camera domain, which is tedious to compile. To address this issue, we present a novel few-shot domain adaptation method to utilize the existing source camera labeled data with few labeled samples from the target camera to improve the target domain's enhancement quality in extreme low-light imaging. Our experiments show that only ten or fewer labeled samples from the target camera domain are sufficient to achieve similar or better enhancement performance than training a model with a large labeled target camera dataset. To support research in this direction, we also present a new low-light raw image dataset captured with a Nikon camera, comprising short-exposure and their corresponding long-exposure ground truth images.

PDF Abstract


Introduced in the Paper:

Nikon RAW Low Light

Used in the Paper:

SID Canon RAW Low Light
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Low-Light Image Enhancement Canon RAW Low Light FSDA-LL PSNR 33.22 # 1
SSIM 0.896 # 1
Domain Adaptation Canon RAW Low Light FSDA-LL Sony -> Canon PSNR 33.22 # 1
SSIM 0.896 # 1
Low-Light Image Enhancement Nikon RAW Low Light FSDAL-LL PSNR 30.3 # 1
SSIM 0.913 # 1
Domain Adaptation Nikon RAW Low Light FSDA-LL Sony -> Nikon PSNR 30.3 # 1
SSIM 0.913 # 1


No methods listed for this paper. Add relevant methods here