Skip to content Skip to navigation
University of Warwick
  • Study
  • |
  • Research
  • |
  • Business
  • |
  • Alumni
  • |
  • News
  • |
  • About

University of Warwick
Publications service & WRAP

Highlight your research

  • WRAP
    • Home
    • Search WRAP
    • Browse by Warwick Author
    • Browse WRAP by Year
    • Browse WRAP by Subject
    • Browse WRAP by Department
    • Browse WRAP by Funder
    • Browse Theses by Department
  • Publications Service
    • Home
    • Search Publications Service
    • Browse by Warwick Author
    • Browse Publications service by Year
    • Browse Publications service by Subject
    • Browse Publications service by Department
    • Browse Publications service by Funder
  • Help & Advice
University of Warwick

The Library

  • Login
  • Admin

A machine learning driven sky model

Tools
- Tools
+ Tools

Satilmis, Pinar, Bashford-Rogers, Thomas, Chalmers, Alan and Debattista, Kurt (2017) A machine learning driven sky model. IEEE Computer Graphics and Applications, 37 (1). pp. 80-91. doi:10.1109/MCG.2016.67

[img]
Preview
PDF
WRAP_mldsm_pinar_s.pdf - Accepted Version - Requires a PDF viewer.

Download (11Mb) | Preview
Official URL: http://dx.doi.org/10.1109/MCG.2016.67

Request Changes to record.

Abstract

Sky illumination is important for generating realistic renderings of virtual environments in a number of applications ranging from entertainment to archaeology. Current solutions use complex analytical models which can be costly to compute interactively; or require the capture of sky environment maps which constitute a laborious and impractical task in order to obtain smooth animations. In this work, we present an alternative model for sky illumination based on machine learning. This approach compactly represents sky illumination from both existing analytic sky models and from captured environment maps. For analytic models, our approach leads to a low, constant runtime cost for evaluating lighting. When applied to environment maps, our approach approximates the captured lighting at a significantly reduced memory cost, and enables smooth transitions of sky lighting to be created from a small set of environment maps captured at discrete times of day. This makes capture and rendering of real world sky illumination a practical proposition. Our method encodes the non-linear mapping of sun and view direction to radiance values using a single layer Artificial Neural Network. The network is trained using a sparse set of samples which capture the properties of the lighting at various sun positions. Results demonstrate accuracy close to the ground truth for both analytical and capture based methods. Our approach has a low runtime overhead meaning that it can be used as a generic approach for both offline and real-time applications.

Item Type: Journal Article
Subjects: Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software
Divisions: Faculty of Science > WMG (Formerly the Warwick Manufacturing Group)
Library of Congress Subject Headings (LCSH): Computer graphics, Machine learning, Neural circuitry
Journal or Publication Title: IEEE Computer Graphics and Applications
Publisher: Institute of Electrical and Electronics Engineers
ISSN: 0272-1716
Official Date: January 2017
Dates:
DateEvent
January 2017Published
25 May 2016Available
2015UNSPECIFIED
Volume: 37
Number: 1
Number of Pages: 9
Page Range: pp. 80-91
DOI: 10.1109/MCG.2016.67
Status: Peer Reviewed
Publication Status: Published
Access rights to Published version: Restricted or Subscription Access

Request changes or add full text files to a record

Repository staff actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics

twitter

Email us: wrap@warwick.ac.uk
Contact Details
About Us