“Automated light probes from capture to render for Peter Rabbit”
Conference:
Type(s):
Interest Area:
- Rendering
Entry Number: 17
Title:
- Automated light probes from capture to render for Peter Rabbit
Presenter(s)/Author(s):
Abstract:
We created an efficient pipeline for automated, HDR light probes for the hybrid live-action / animated feature film Peter Rabbit. A specially developed “360°” spherical camera allows on-set acquisition at more positions and in less time than traditional techniques. Reduced capture time, drastically simplified stitching and a custom multiple-exposure raw to HDR process minimizes artefacts in the resulting images. A semi-automated system recovers clipped radiance in direct sunlight using surfaces with known properties. By recording capture location and orientation and combining with other scene data we produce automated rendering setups using the light probes for illumination and projection onto 3d render geometry.
References:
Procedural Photograph Generation from Actual Gameplay: Snapshot AI in FINAL FANTASY XV INDIECAM. 2016. INDIECAM nakedEYE Product Description. http://www.indiecam.com/products/nakedeye/. (2016).Google Scholar
Jessi Stumpfel. 2004. HDR Lighting Capture of the Sky and Sun. Master’s thesis. California Institute of Technology.Google Scholar
Alex Suter, Dan Lobl, Victor Schutz, and Brian Gee. 2016. LightCraft: Extract Light Position and Information from On-set Photography. In ACM SIGGRAPH 2016 Talks (SIGGRAPH ’16). ACM, New York, NY, USA, Article 15, 2 pages. Google Scholar
