Learning Lightprobes for Mixed Reality Illumination

David Mandl*, Kwang Moo Yi, Peter Mohr-Ziak, Peter M. Roth, Pascal Fua, Vincent Lepetit, Dieter Schmalstieg, Denis Kalkofen

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review


This paper presents the first photometric registration pipeline for
Mixed Reality based on high quality illumination estimation using
convolutional neural networks (CNNs). For easy adaptation and deployment
of the system, we train the CNNs using purely synthetic
images and apply them to real image data. To keep the pipeline accurate
and efficient, we propose to fuse the light estimation results
from multiple CNN instances and show an approach for caching
estimates over time. For optimal performance, we furthermore explore
multiple strategies for the CNN training. Experimental results
show that the proposed method yields highly accurate estimates for
photo-realistic augmentations.
Original languageEnglish
Title of host publicationProceedings of the 2017 IEEE International Symposium on Mixed and Augmented Reality, ISMAR 2017
PublisherInstitute of Electrical and Electronics Engineers
Pages82 - 89
Number of pages8
ISBN (Electronic)978-153862943-7
Publication statusPublished - 9 Oct 2017
Event2017 IEEE International Symposium on Mixed and Augmented Reality: ISMAR 2017 - La Cité, Nantes, France
Duration: 9 Oct 201713 Oct 2017


Conference2017 IEEE International Symposium on Mixed and Augmented Reality
Abbreviated titleISMAR 2017
Internet address


Dive into the research topics of 'Learning Lightprobes for Mixed Reality Illumination'. Together they form a unique fingerprint.

Cite this