Flocks, herds and schools: A distributed behavioral model
SIGGRAPH '87 Proceedings of the 14th annual conference on Computer graphics and interactive techniques
Resolving Motion Correspondence for Densely Moving Points
IEEE Transactions on Pattern Analysis and Machine Intelligence
Tracking and labelling of interacting multiple targets
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part III
| Hi-index | 0.00 |
In this paper we propose a novel method to obtain 3D motion trajectories of dense particle swarms using multiple cameras, which facilitates the study of animal grouping behavior. The proposed method aims at minimizing the two kinds of ambiguities: stereo matching ambiguity and motion correspondence ambiguity. We introduce Verification View which provides additional epipolar constraint to significantly reduce stereo matching ambiguity. The proposed method employs optimal assignment with state prediction and candidate filtering to establish temporal association of particles. The performance on simulated particle swarms demonstrates the superiority of our methods in reducing the two kinds of ambiguities, compared with state-of-the-art. Besides, the proposed method successfully reconstructs hundreds of trajectories of Drosophila in real-world experiment, in only a few seconds.