Department of Pattern Recognition
old web
ÚTIA
AV ČR
Log in
Search this site:
Home
Seminars
People
Demos
Lectures / Tutorials
Resources / Software
Publications
Topics (57)
3D Data Modelling (13)
3D Objects Measurement (2)
Bidirectional Texture Function Modelling (7)
BTF Measurement (1)
Virtual Reality (3)
Benchmarking (2)
Classification (4)
Content Based Image Retrieval (CBIR) (1)
Dynamic Textures (3)
Colour Dynamic Textures Synthesis (1)
Dynamic Texture Modelling (1)
Illumination and Rotation Invariance (1)
Image & Video Analysis / Processing (11)
Image Retrieval (2)
Image Segmentation (5)
Image Sequence Restoration (2)
Remote Sensing (1)
Texture Synthesis (1)
Machine Learning (11)
Mixture Models (11)
Application of Distribution Mixtures (3)
Feature Selection (4)
Mammogram Enhancement (2)
Mathematical Image Modelling (1)
Medical Image Recognition (1)
Pattern Recognition (1)
Probabilistic Expert Systems (1)
Probabilistic Neural Networks (2)
Probabilistic Texture Synthesis (5)
Sampling Based Texture Synthesis (2)
Spatial Data Modelling (1)
Statistical Pattern Recognition (8)
Document Classification (1)
Textural Features (1)
Texture Editing (1)
Texture Segmentation (2)
Visual quality measures (1)
Home
›
Topics
›
Sampling Based Texture Synthesis
A Roller - Fast Sampling-Based Texture Synthesis Algorithm
Michal Haindl
Martin Hatka
Abstract:
A method for synthesizing natural textures that realistically matches given colour texture appearance. The novel texture synthesis method, which we call the roller, is based on the overlapping tiling and subsequent minimum error boundary cut. An optimal double toroidal patch is seamlessly repeated during the synthesis step. While the method allows only moderate texture compression it is extremely fast and easily implementable in a graphical hardware for purpose of real-time rendering.
Extremely fast method, there is no computation in synthesis.
Unlimited seamles BTF texture enlargement.
Easy GPU implementation.
Mild BTF texture compression ratio (1:3), occasional failure on textures with dominant directional low frequencies.
Demo
Results
input texture
enlarged texture
double-toroidal tile
Reference:
Haindl, M.
, and
M. Hatka
,
"
A Roller - Fast Sampling-Based Texture Synthesis Algorithm
",
The 13th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2005
, Plzen, University of Western Bohemia, pp. 80-83, February, 2005.
BibTex
Google Scholar
Haindl, M.
, and
M. Hatka
,
"
BTF Roller
",
Texture 2005: Proceedings of 4th Internatinal Workshop on Texture Analysis and Synthesis
, Edinburgh, Heriot-Watt University, pp. 89-94, October, 2005.
BibTex
Google Scholar