MSR-Net: Multi-Scale Relighting Network for One-to-One Relighting

• DOI: 10.48550/arxiv.2107.06125
• Publication Date: 2021-01-01
• Authors: Sourya Dipta Das; Nisarg A. Shah; Saikat Dutta
• Keywords: FOS: Computer and information sciences; Computer Vision and Pattern Recognition (cs.CV); Computer Science - Computer Vision and Pattern Recognition; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

Deep image relighting allows photo enhancement by illumination-specific retouching without human effort and so it is getting much interest lately. Most of the existing popular methods available for are run-time intensive memory inefficient. Keeping these issues in mind, we propose use Stacked Multi-Scale Hierarchical Network, which aggregates features from each at different scales. Our solution differentiable robust translating illumination setting input to target image. Additionally, have also shown that using a multi-step training approach this problem with two loss functions can significantly boost performance achieve high quality reconstruction relighted