Sciweavers

ICPR
2010
IEEE

Shape Guided Maximally Stable Extremal Region (MSER) Tracking

13 years 5 months ago
Shape Guided Maximally Stable Extremal Region (MSER) Tracking
Maximally Stable Extremal Regions (MSERs) are one of the most prominent interest region detectors in computer vision due to their powerful properties and low computational demands. In general MSERs are detected in single images, but given image sequences as input, the repeatability of MSER detection can be improved by exploiting correspondences between subsequent frames by feature based analysis. Such an approach fails during fast movements, in heavily cluttered scenes and in images containing several similar sized regions because of the simple feature based analysis. In this paper we propose an extension of MSER tracking by considering shape similarity as strong cue for defining the frame-to-frame correspondences. Efficient calculation of shape similarity scores ensures that realtime capability is maintained. Experimental evaluation demonstrates improved repeatability and an application for tracking weakly textured, planar objects.
Michael Donoser, Hayko Riemenschneider, Horst Bisc
Added 12 Oct 2010
Updated 12 Oct 2010
Type Conference
Year 2010
Where ICPR
Authors Michael Donoser, Hayko Riemenschneider, Horst Bischof
Comments (0)