Steerable Texture Synthesis for Vector Field Visualization
| dc.contributor.author | Taponecco, Francesca | en_US |
| dc.coverage.spatial | Darmstadt | en_US |
| dc.date.accessioned | 2015-01-21T06:45:21Z | |
| dc.date.available | 2015-01-21T06:45:21Z | |
| dc.date.issued | Oct 2006 | en_US |
| dc.description.abstract | Visualization is a fundamental field of research with uncountable applications, spanning from the fields of computer graphics and vision to the humanities. A key feature in visualization studies is thus the interdisciplinary nature of this research field, and it is interesting to note the numerous relative benefits and open directions where investigations can be guided. In the last years, importance of visualization is constantly growing as, thanks to the fast computer advances, it is possible to collect and handle large data sets. Consequently, focusing on scientific visualization techniques still reserves much attention and, although several valid visualization methods exist, further investigation is required. A fundamental open issue is the need for more control. The broad variety of tasks and the different level of expertise of users also require degrees of freedom and adaptivity to allow customizing the visualization process, effectively representing data sets. Such features would be especially beneficial in computer vision and imaging applications as well as for textures generation. The need for local control and the ability to constrain the synthesis of textures are nowadays relevant issues due to the fundamental role that textures play in computer graphics, providing realism and variety in digital scenes and objects. Unfortunately, most synthesis approaches still just allow generating simple homogeneous and regular textures, leaving a great part of texture potential unexplored. In this work, I propose novel techniques for the visualization of vectorial data sets, offering local as well as global control in the visualization process. I use statistical theory from texture synthesis and concepts from perception and cognition to optimize the resulting image and encode the information in the visualization. Furthermore, I introduce straightforward extensions to standard texture synthesis algorithms, allowing the generation of constrained textures, field-driven textures and a variety of texture filtering and transformation effects. | en_US |
| dc.description.seriesinformation | EG Graphics Dissertation Online | |
| dc.format | application/pdf | en_US |
| dc.identifier.doi | 10.2312/diss.20068185 | |
| dc.identifier.uri | https://diglib.eg.org/handle/10.2312/8185 | |
| dc.identifier.uri | https://doi.org/10.2312/diss.20068185 | |
| dc.language | English | en_US |
| dc.publisher | Taponecco, Francesca | en_US |
| dc.title | Steerable Texture Synthesis for Vector Field Visualization | en_US |
| dc.type | Text.PhDThesis | en_US |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- taponecco.pdf
- Size:
- 13.03 MB
- Format:
- Adobe Portable Document Format