Physically inspired dense fusion networks for relighting

Amirsaeed Yazdani, Tiantong Guo, Vishal Monga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

Image relighting has emerged as a problem of significant research interest inspired by augmented reality applications. Physics-based traditional methods, as well as black box deep learning models, have been developed. The existing deep networks have exploited training to achieve a new state of the art; however, they may perform poorly when training is limited or does not represent problem phenomenology, such as the addition or removal of dense shadows. We propose a model which enriches neural networks with physical insight. More precisely, our method generates the relighted image with new illumination settings via two different strategies and subsequently fuses them using a weight map (w). In the first strategy, our model predicts the material reflectance parameters (albedo) and illumination/geometry parameters of the scene (shading) for the relit image (we refer to this strategy as intrinsic image de-composition (IID)). The second strategy is solely based on the black box approach, where the model optimizes its weights based on the ground-truth images and the loss terms in the training stage and generates the relit output directly (we refer to this strategy as direct). While our proposed method applies to both one-to-one and any-to-any relighting problems, for each case we introduce problem-specific components that enrich the model performance: 1) For one-to-one relighting we incorporate normal vectors of the surfaces in the scene to adjust gloss and shadows accordingly in the image. 2) For any-to-any relighting, we propose an additional multiscale block to the architecture to enhance feature extraction. Experimental results on the VIDIT 2020 and the VIDIT 2021 dataset (used in the NTIRE 2021 relighting challenge) reveals that our proposal can outperform many state-of-the-art methods in terms of well-known fidelity metrics and perceptual loss.

Original languageEnglish (US)
Title of host publicationProceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
PublisherIEEE Computer Society
Pages497-506
Number of pages10
ISBN (Electronic)9781665448994
DOIs
StatePublished - Jun 2021
Event2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 - Virtual, Online, United States
Duration: Jun 19 2021Jun 25 2021

Publication series

NameIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (Print)2160-7508
ISSN (Electronic)2160-7516

Conference

Conference2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
Country/TerritoryUnited States
CityVirtual, Online
Period6/19/216/25/21

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Physically inspired dense fusion networks for relighting'. Together they form a unique fingerprint.

Cite this