DSRN: an Efficient Deep Network for Image Relighting

TL;DR

This paper introduces DSRN, a lightweight, real-time deep learning model for image relighting that efficiently adjusts illumination and color temperature, outperforming existing methods in speed and robustness.

Contribution

The paper presents DSRN, a novel, efficient deep network architecture for image relighting that is faster and more memory-efficient than prior models, with improved robustness.

Findings

DSRN is lightweight at 42 MB and achieves 0.0116s inference time for 1024x1024 images.

The model effectively translates image color temperature to target lighting conditions.

Using images illuminated from opposite directions enhances qualitative relighting results.

Abstract

Custom and natural lighting conditions can be emulated in images of the scene during post-editing. Extraordinary capabilities of the deep learning framework can be utilized for such purpose. Deep image relighting allows automatic photo enhancement by illumination-specific retouching. Most of the state-of-the-art methods for relighting are run-time intensive and memory inefficient. In this paper, we propose an efficient, real-time framework Deep Stacked Relighting Network (DSRN) for image relighting by utilizing the aggregated features from input image at different scales. Our model is very lightweight with total size of about 42 MB and has an average inference time of about 0.0116s for image of resolution $1024 \times 1024$ which is faster as compared to other multi-scale models. Our solution is quite robust for translating image color temperature from input image to target image and also performs moderately for light gradient generation with respect to the target image. Additionally, we show that if images illuminated from opposite directions are used as input, the qualitative results improve over using a single input image.