550 likes | 1.1k Views
Manifold Bootstrapping for SVBRDF Capture. Yue Dong, Jiaping Wang, Xin Tong, John Snyder, Moshe Ben-Ezra, Yanxiang Lan, Baining Guo Tsinghua University Microsoft Research Asia Microsoft Research. High-Quality SVBRDF Acquisition. high spatial variation.
E N D
Manifold Bootstrapping for SVBRDF Capture Yue Dong, Jiaping Wang, Xin Tong, John Snyder, Moshe Ben-Ezra, Yanxiang Lan, Baining Guo Tsinghua University Microsoft Research Asia Microsoft Research
High-Quality SVBRDF Acquisition high spatial variation high angular variation fast and simple too!
Related Work • brute force (6D) measurement (gonioreflectometer) [Dana et al. 1999, McAllister et al. 2002, Lawrence et al. 2006] • slow • expensive, specialized rig
Related Work • single-pass fitting methods [Lensch et al. 2003, Goldman et al. 2005, Zickler et al. 2005] • measures large dataset • fits limited models(parametric/isotropic)
Related Work • two-pass methods • linearly combine two representatives based on diffuse color[Debevec et al. 2003]
Related Work • two-pass methods • linearly combine two representatives based on diffuse color[Debevec et al. 2003]
Related Work • two-pass methods • linearly combine two representatives based on diffuse color[Debevec et al. 2003] • use existing BRDF database of representatives: non-specialized and isotropic[Matusik et al. 2003b; Weyrich 2006]
Observation • BRDF spatial variation is complex: • tangent/normal/local frame rotates • specularity/anisotropy varies • specular lobe’s falloff and cross-section changes • forms low-dimensional manifold over given target. • manifold isn’t globally linear[Matusik et al. 2003a] • manifold is locally linear.
SVBRDF Manifold locally linear globally non-linear
Local vs. Global Interpolation local interpolation
Local vs. Global Interpolation global interpolation
SVBRDF Manifold Bootstrapping SVBRDF Manifold Representative Space
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements Material Sample
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements Key Measurements Material Sample
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements every pixel Key Space Key Measurements
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements x Local EmbeddingIn Key Space Key Space Key Measurements
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements x Local EmbeddingIn Key Space Key Measurements
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements Local Embedding of x In Representative Space x Local EmbeddingIn Key Space Key Measurements
SVBRDF Manifold Bootstrapping Representative Space Representative Measurements Reconstructed BRDF of x Local Embedding of x In Representative Space x Local EmbeddingIn Key Space Key Measurements
Results Real Material Sample
Outline • Data Acquisition • SVBRDF Reconstruction • Validation
Representative BRDFs • portable BRDF scanner • 6 LED light directions, 320x240 view directions • data amplification by microfacet model • 0.1s per BRDF
Key Measurements • fixed camera • background environmental lighting + moving area source
Timing • representative BRDFs and key measurements • 10-15 minutes • data processing • less than 5 minutes
Outline • Data Acquisition • SVBRDF Reconstruction • Validation
SVBRDF Reconstruction Representative BRDFs
Representative Local Interpolation ? + w3 = w1 + w2 BRDF of x x Representative BRDFs Material Sample
Representative Local Interpolation • choose which representatives to interpolate from • solve for weights wi ? w3 = w1 w1 + w2 + w3 w2 BRDF of x x Representative BRDFs Material Sample
Key Measurement Environment Lighting Representative BRDFs Projected Keys of Representative BRDFs Key Measurements Material Sample
Key Measurement Projected Keys of Representative BRDFs Key Measurements
Key Local Interpolation Key of x nearest neighbor in key space x Projected Keys of Representative BRDFs Key Measurements
Key Local Interpolation • solve for weights: LLE[Roweis & Saul 2000] where Key of x x + w3 = w1 + w2 Key Measurements
BRDF Reconstruction Neighborhood Key of x + w3 = w1 + w2 Local Embedding in Key Space
BRDF Reconstruction + w3 + w2 = w1 Key of x weights BRDF of x + w3 = w1 + w2 Local Embedding in Key Space
Outline • Data Acquisition • SVBRDF Reconstruction • Validation
Key Space vs. Representative Space • Projection depend on the environmental lighting conditions • preserve distances ⇒ preserve BRDF manifold structure
Key Space vs. Representative Space • Projection depend on the environmental lighting conditions • preserve distances ⇒ preserve BRDF manifold structure • global distances ⇒ preserve neighborhoods • local distances ⇒ preserve weights
Distance Preservation • preservation evaluation
Distance Preservation • preservation evaluation • # of lighting conditions
Distance Preservation • preservation evaluation • # of lighting conditions • criterion:global:τg> 0.9local:τl> 0.85
Results Real Material Sample
Extension to local frame variations • Normal variations • Tangent rotations
Representative Enlargement … enlarged BRDFs over normal rotation … enlarged BRDFs over tangent rotation
Results Real Material Sample
Results Real Material Sample
Conclusion • Manifold bootstrapping captures high-resolution SVBRDF • assumes BRDF forms low-dimensional manifold • decomposes acquisition into two phases • makes sparse measurement in both • phase one (representatives) = sparse spatial, dense angular • phase two (keys) = sparse angular, dense spatial • simplifies and accelerates the capture process
Conclusion • Manifold bootstrapping captures high-resolution SVBRDF • assumes BRDF forms low-dimensional manifold • decomposes acquisition into two phases • makes sparse measurement in both • phase one (representatives) = sparse spatial, dense angular • phase two (keys) = sparse angular, dense spatial • simplifies and accelerates the capture process
Acknowledgements • Paul Debevec for HDR images • Steve Lin for video narration • Anonymous reviewers for helpful comments