350 likes | 518 Views
Fast Texture Synthesis using Tree-structured Vector Quantization. Li-Yi Wei Marc Levoy. Computer Graphics Group Stanford University. Introduction. Texture Synthesis. Input. Result. Desirable Properties. Result looks like the input Efficient General Easy to use
E N D
Fast Texture Synthesis using Tree-structured Vector Quantization Li-Yi Wei Marc Levoy Computer Graphics Group Stanford University
Introduction Texture Synthesis Input Result
Desirable Properties • Result looks like the input • Efficient • General • Easy to use • Extensible
Previous Work • Procedural Synthesis • Perlin 85, Witkin 91, Worley 96 • Statistical Feature Matching • Heeger 95, De Bonet 97, Simoncelli 98 • Markov Random Fields • Popat 93, Efros 99
Outline • Basic algorithm • Multi-resolution algorithm • Acceleration • Applications
Texture Model • Textures are • local • stationary • Model textures by • local spatial neighborhoods
Basic Algorithm • Exhaustively search neighborhoods
Noise Neighborhood • Use causal neighborhoods Input Causal Non-causal
Neighborhood • Neighborhood size determines the quality & cost 33 55 77 739 s 423 s 528 s 1111 4141 99 24350 s 1020 s 1445 s
High resolution Low resolution Multi-resolution Pyramid
Benefit • Better image quality & faster computation 1 level 55 1 level 1111 3 levels 55
Results Oriented Random Regular Semi-regular
Failures • Non-planar structures • Global information
Comparison Input 12 secs Heeger 95 De Bonet 97 Efros 99 Our method 503 secs 1941 secs
Acceleration • Computation bottleneck: neighborhood search
1 2 3 4 5 6 7 8 9 10 11 12 High dimensional point/vector 1 2 3 4 5 6 7 8 9 10 11 12 Nearest Point Search • Treat neighborhoods as high dimensional points Neighborhood
Acceleration • Nearest point search in high dimensions • [Nene 97] • Cluster-based model for textures • [Popat 93] • Tree-structured Vector Quantization • [Gersho 92]
Timing • Time complexity : O(log N) instead of O(N) • 2 orders of magnitude speedup for non-trivial images Efros 99 Full searching TSVQ 1941 secs 503 secs 12 secs
Results: Brodatz Textures D103 D20 Input Exhaustive: 360 secs TSVQ: 7.5 secs
Possible Solution • Multi-resolution blending [Burt & Adelson 83] • produce visible boundaries
Possible Solution • Original raster-scan algorithm • discontinuities at right and bottom boundaries
Possible Solution • Adaptive neighborhoods [Efros 99] • Hard to accelerate
Modifications • Need to use a single symmetric neighborhood • 2 pass algorithm with extrapolation • Spiral order synthesis
Result • Extrapolation ? ? ? ?
Result • Image editing by texture replacement
Application 2:Temporal Texture • Indeterminate motions both in space and time • fire, smoke, ocean waves • How to synthesize? • extend our 2D algorithm to 3D
Temporal Texture Input Result Fire Smoke Waves
Future Work • More general “textures” • light fields, solid textures • motion signals • displacement maps • Real time texture synthesis
Kris Popat Alyosha Efros Stanford Graphics Group Intel, Interval, Sony Acknowledgment More information http://graphics.stanford.edu/projects/texture/