[[Editing a VisnetWiki Page]]

Editing a VisnetWiki Page

Syntax instructions for a dokuwiki can be found .

Equations have to be instered as images at the moment.

Images can be inserted using the in the on the editing page. Once you have uploaded your image you can include it by double clicking on its icon in the Media Manager or include it in the text with double curly brackets. More syntax on images can be found .

About VISNET Wiki

There are 3 major VISNET related topics:

  • Video Coding
  • Audiovisual Media Processing
  • Security

Please stick to this structure when you create a new subtopic. Subtopics at the first level should be general topics fitting into one of the three major VISNET topics such as e.g.

  • Video Coding
  • Audiovisual Media Processing
    • Audio Analysis
    • Human Analysis
    • Video Segmentation and Tracking
  • Security

More detailed articles related to VISNET research activities or other research results should be discussed in seperate articles integrated into the major subtopics. These articles can be e.g. linked to in a section “Methods” or “Related topics” in the general topics. Please try to integrate your article into one of the main topics or in one of the subtopics that already exist or create a general subtopic yourself in that you can intergrate your article.

The structure of each page should include the main article, and e.g. the sections “Related Topics”, “References” and “Links”, and link already existing related articels in “Related Topics”, like the following short example. Please try avoid VISNET internal structure in order to promote external collaboration.

Conferences related to Visnet research topics can be added in the Conference section on the welcome page.

Wiki pages containing listing of databases can be linked to in the databases section to have quick access to database lists from the welcome page.

Example: Face Recognition and Analysis

Face recognition includes a set of challenges like expression variations, occlusions of facial parts, similar identities, resolution of the acquired images, aging of the subjects and many others. Among all these challenges, most of the face recognition techniques have evolved in order to overcome two main problems: illumination and pose variation as described by Phillips et al. [1] as well as Zhao and Chellapa [2]. Either of these influences can cause serious performance degradation in a face recognition system. Pose can change the appearance of an object drastically, and in the most of the cases these differences induced by pose variations are larger than differences between individuals, what makes the recognition task difficult. The same statement is valid for illumination variation. Therefore, pose and illumination (among other challenges) are the main causes for the degradation of 2D face recognition algorithms.

A recent public test at a German central train station showed only a recognition rate of 30% with a training set of 200 people out of 23000 people as described in the report of Pretzel and Lotz [3].


  • Pose Invariant Face Recognition

Related Topics

  • Surveillance
  • Principle Component analysis


[1] Phillips, P.J., Grother, P., Micheals, R., Blackburn, D., Tabassi, E., Bone, J.: Face Recognition Vendor Test 2002: Evaluation Report. Technical Report NISTIR 6965, National Institute of Standards and Technology (2003)

[2] Zhao, W., Chellapa, R.: Face Processing: Advanced modeling and methods. Academic Press (2006)

[3] Pretzel, Lotz: Research project: Face recognition as a search tool. Technical report, Bundeskriminalamt Wiesbaden (2007)


editing.txt · Last modified: 2008/07/03 10:37 by admuser
Recent changes RSS feed Creative Commons License Powered by PHP Valid XHTML 1.0 Valid CSS