Abstract | ||
---|---|---|
We describe a method for tracking animals in wildlife footage. It uses a CONDENSATIONparticle filtering frame- work driven by learnt characteristics of specific animals. The key contribution is a periodic model of animal motion based on the relative positions over time of trackable fea- tures at significant body points. We also introduce tech- niques for maintaining a multimodal state density within the particle filter over time to enable consistent tracking of mul- tiple animals. Initial experiments show that the approach has considerable potential. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1109/ICPR.2002.1048227 | Pattern Recognition, 2002. Proceedings. 16th International Conference |
Keywords | DocType | Volume |
filtering theory,image motion analysis,image sequences,object detection,optical tracking,video signal processing,condensation particle filtering framework,animal characteristics,body points,multimodal state density,multiple animal tracking,periodic animal motion model,trackable features,video,wildlife footage | Conference | 2 |
ISSN | Citations | PageRank |
1051-4651 | 5 | 1.02 |
References | Authors | |
4 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
David Tweed | 1 | 60 | 7.71 |
Andrew Calway | 2 | 645 | 54.66 |