DocumentCode :
2485107
Title :
Online object recognition by MSER trajectories
Author :
Riemenschneider, Hayko ; Donoser, Michael ; Bischof, Horst
Author_Institution :
Inst. for Comput. Graphics & Vision, Graz Univ. of Technol., Graz
fYear :
2008
fDate :
8-11 Dec. 2008
Firstpage :
1
Lastpage :
4
Abstract :
This work presents a robust online learning and recognition system. The basic idea is to exploit information from tracking an object during the recognition and/or learning stage to obtain increased robustness and better recognition results. Object tracking by means of an extended MSER tracker is utilized to detect local features and construct their trajectories. Compact object representations are formed by summarizing the trajectories. All steps are performed online including the MSER detection, tracking, summarization, SIFT description as well as learning and recognition based on a vocabulary tree. The proposed method is evaluated on realistic video sequences which prove the increased performance for robust online recognition.
Keywords :
feature extraction; learning (artificial intelligence); object recognition; MSER tracker; MSER trajectories; SIFT description; local feature detection; object representations; object tracking; online learning; online object recognition; realistic video sequences; recognition system; robust online recognition; Computer graphics; Computer vision; Object detection; Object recognition; Robustness; Tracking; Training data; Trajectory; Video sequences; Vocabulary;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 2008. ICPR 2008. 19th International Conference on
Conference_Location :
Tampa, FL
ISSN :
1051-4651
Print_ISBN :
978-1-4244-2174-9
Electronic_ISBN :
1051-4651
Type :
conf
DOI :
10.1109/ICPR.2008.4761604
Filename :
4761604
Link To Document :
بازگشت