Animation of reactive gaseous fluids through chemical kinetics
SCA '04 Proceedings of the 2004 ACM SIGGRAPH/Eurographics symposium on Computer animation
Image-based tomographic reconstruction of flames
SCA '04 Proceedings of the 2004 ACM SIGGRAPH/Eurographics symposium on Computer animation
Enhanced illumination of reconstructed dynamic environments using a real-time flame model
AFRIGRAPH '06 Proceedings of the 4th international conference on Computer graphics, virtual reality, visualisation and interaction in Africa
Boundary matting for view synthesis
Computer Vision and Image Understanding
Image-based tomographic reconstruction of flames
SIGGRAPH '04 ACM SIGGRAPH 2004 Sketches
Adaptive grid optical tomography
Graphical Models - Special issue on the vision, video and graphics conference 2005
Density estimation for dynamic volumes
Computers and Graphics
Weighted Minimal Hypersurface Reconstruction
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Theory of Refractive and Specular 3D Shape by Light-Path Triangulation
International Journal of Computer Vision
Physically-based realistic fire rendering
NPH'06 Proceedings of the Second Eurographics conference on Natural Phenomena
Real-time rendering and animation of plentiful flames
NPH'07 Proceedings of the Third Eurographics conference on Natural Phenomena
Real-time animation of various flame shapes
VAST'06 Proceedings of the 7th International conference on Virtual Reality, Archaeology and Intelligent Cultural Heritage
Volumetric reconstruction, compression and rendering of natural phenomena from multi-video data
VG'05 Proceedings of the Fourth Eurographics / IEEE VGTC conference on Volume Graphics
| Hi-index | 0.00 |
This paper considers the problem of reconstructing visuallyrealistic 3D models of fire from a very small setof simultaneous views (even two). By modeling fire as asemi-transparent 3D density field, we show that fire reconstructionis equivalent to a severely under-constrained computerizedtomography problem, for which traditional methodsbreak down. Our approach is based on the observationthat every pair of photographs of a semi-transparentscene defines a unique density field, called a Flame Sheet,that (1) concentrates all its density on one connected, semi-transparentsurface, (2) reproduces the two photos exactly,and (3) is the most spatially-coherent density field that doesso. From this observation, we reduce fire reconstruction tothe convex combination of sheet-like density fields, each ofwhich is derived from the Flame Sheet of two input photos.Experimental results suggest that this method enables high-qualityview extrapolation without over-fitting artifacts.