PERTEX · Perceptual analysis and measurement of surface texture
FP6 — Marie Curie Actions (Human Resources and Mobility)
- Duration
- 2007-04-01 → 2009-03-31
- EU contribution
- €168,798
- Participants
- 1
- Scheme
- EIF
Lines connect the coordinator with its partners.
Results in brief
Final Activity Report Summary - PerTex (Perceptual Analysis and Measurement of Surface Texture)
One of the most challenging technological problems that are faced by the film and computer games industries is the production of photorealistic imagery of surface materials such as stone, wood, leather, plastic, metal etc. Furthermore, certain applications such as automotive or architectural interior design actually need additional visual accuracy to ensure that the graphics reproduce completely the visual perceptions invoked by the original materials. This requires processing of very large datasets involving significant computational demands. This project addressed these problems by exploiting the tools and techniques of cognitive psychology, computer vision and computer graphics to solve the riddle of the kind of data that was important to the human visual system when rendering real surface materials. This cross-disciplinary approach helped us to understand the way people observed and evaluated the appearance of real-world materials as well as to significantly improve the speed and quality of computer graphics reproduction without compromising visual accuracy.
Data: CORDIS, © European Union
Project objective
In this project, we aim to identify perceptual dimensions of surface texture and to find new computational features corresponding to these dimensions. By surface texture we mean the complex surface characteristics regarded as a combination of geometry (surface relief) and variation of reflectance function. Computational features are statistical descriptors of surface textures. We believe that this problem can be characterised as follows:1. Importance: Surface appearance is a critical characteristic of nearly all of today¿s commodity items. However, there are currently no accepted ways for producing perceptually meaningful measurements of surface texture, or even an agreement on what the principal perceptual dimensions are. Yet, this ability is key to producing intuitive and effective information systems that will be able to manage the explosion of this type of data that will soon take place due to recent advances in computer graphics and surface capture techniques.2. Novelty: Psychometric studies into texture have so far been restricted to using single, still images of surface texture. We believe that such data provide insufficient stimuli for observers and that they bias the computational features in often uncontrolled or unknown ways. We will use real time photorealistic rendering of BTF (Bidirectional Texture Functions) to provide observers with the rich stimuli they normally experience when examining surface textures and we will derive computational features that are intrinsic to surface directly from BTF. We are not aware of this approach having been reported in the literature before.
Original text from CORDIS.
Participants
- HERIOT-WATT UNIVERSITY · EDINBURGHCoordinatorUnited Kingdom
Links
Data: CORDIS, © European Union