|
 |
|
Technical Papers
Materials and Images
Thursday, 29 November 16:15 - 18:15 | Peridot 206
 |
Material Memex: Automatic Material Suggestions for 3D ObjectsMaterial found on 3D objects and their parts correlates with the geometric shape of the parts and their relation to other parts of the same object. This work proposes to model this context-dependent correlation by learning it from a database containing several hundreds of objects and their materials. Arjun Jain, Max-Planck-Institut für Informatik |
 |
Interactive Bi-scale Editing of Highly Glossy MaterialsWe present a new technique for bi-scale material editing using Kei Iwasaki, Wakayama University |
 |
An Inverse Problem Approach for Automatically Adjusting the Parameters for Rendering Clouds Using PhotographsParameters for rendering synthetic clouds realistically are estimated automatically from photographs. Genetic algorithms are used to search for optimal parameters. Once the parameters have been obtained, the user can render the synthetic clouds with various viewpoints, sunlight directions, and sunlight colors. Yoshinori Dobashi, Hokkaido University, JST CREST |
 |
Lighting Hair From The Inside: A Thermal Approach To Hair ReconstructionWe present a novel technique to reconstruct hairstyles based on images taken with a thermal camera, avoiding thus several issues related to conventional image-based approaches. Tomás Lay Herrera, Institut für Informatik II, Universität Bonn |
 |
New Measurements Reveal Weaknesses of Image Quality Metrics in Evaluating Graphics ArtifactsWe propose rendering-oriented datasets for image quality evaluation, which provide detailed distortion maps along with the probability of their detection by human observers. For existing full-reference image quality metrics our datasets turned out to be very demanding, and our analysis of metric failures suggests directions for improvement. Martin Cadik, Max-Planck-Institut für Informatik |
Sponsored by ACMSIGGRAPH