Publication Details
Evaluation and comparison of tracking methods using meeting omnidirectional images
Beran Vítězslav, doc. Ing., Ph.D. (DCGM FIT BUT)
Sumec Stanislav, Ing., Ph.D. (DCGM FIT BUT)
Zemčík Pavel, prof. Dr. Ing. (DCGM FIT BUT)
omnidirectional vision, meeting recordings, tracking, evaluation metods.
Visual cues, such as gesturing, looking at each other or monitoring each others facial expressions, play an important role in meetings. Such information can be used for indexing of multimedia meeting recordings. These situations are strongly focused nowadays. The omnidirectional system usage in such situations brings many advantages as portability, easy installation, large field of view, low cost etc. That is why we choose such scenarios for testing the omnidirectional system. Information about differences between omni-directional and classical images both for human presentation and tracking purposes is needed. We try to compare two different tracking methods on the various video sequences. The results of the tracking methods can help to demonstrate the benefits or drawbacks of the omni-directional system. The evaluation scheme was developed to bring us the aspects which affect the vision algorithms for detection and tracking of human bodies.
@INPROCEEDINGS{FITPUB8423,
author = "Igor Pot\'{u}\v{c}ek and V\'{i}t\v{e}zslav Beran and Stanislav Sumec and Pavel Zem\v{c}\'{i}k",
title = "Evaluation and comparison of tracking methods using meeting omnidirectional images",
pages = 12,
booktitle = "Workshop on Multimodal Interaction and Related Machine Learning Algorithms (MLMI)",
year = 2007,
location = "Brno, CZ",
language = "english",
url = "https://www.fit.vut.cz/research/publication/8423"
}