English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

Consistent Interactive Augmentation of Live Camera Images with Correct Near-field Illumination

MPS-Authors
/persons/resource/persons44531

Grosch,  Thorsten
Computer Graphics, MPI for Informatics, Max Planck Society;

External Resource
No external resources are shared
Fulltext (restricted access)
There are currently no full texts shared for your IP range.
Fulltext (public)
There are no public fulltexts stored in PuRe
Supplementary Material (public)
There is no public supplementary material available
Citation

Grosch, T., Eble, T., & Mueller, S. (2007). Consistent Interactive Augmentation of Live Camera Images with Correct Near-field Illumination. In S. N. Spencer (Ed.), VRST 2007: ACM Symposium on Virtual Reality Software and Technology (pp. 125-132). New York, NY, USA: ACM.


Cite as: https://hdl.handle.net/11858/00-001M-0000-000F-1EB7-5
Abstract
Inserting virtual objects in real camera images with correct lighting is an active area of research. Current methods use a high dynamic range camera with a fish-eye lens to capture the incoming illumination. The main problem with this approach is the limitation to distant illumination. Therefore, the focus of our work is a real-time description of both near - and far-field illumination for interactive movement of virtual objects in the camera image of a real room. The daylight, which is coming in through the windows, produces a spatially varying distribution of indirect light in the room; therefore a near-field description of incoming light is necessary. Our approach is to measure the daylight from outside and to simulate the resulting indirect light in the room. To accomplish this, we develop a special dynamic form of the irradiance volume for real-time updates of indirect light in the room and combine this with importance sampling and shadow maps for light from outside. This separation allows object movements with interactive frame rates (10--17 fps). To verify the correctness of our approach, we compare images of synthetic objects with real objects.