Shape-and-Behavior Encoded Tracking of Bee Dances

TitleShape-and-Behavior Encoded Tracking of Bee Dances
Publication TypeJournal Articles
Year of Publication2008
AuthorsVeeraraghavan A, Chellappa R, Srinivasan M
JournalPattern Analysis and Machine Intelligence, IEEE Transactions on
Volume30
Issue3
Pagination463 - 476
Date Published2008/03//
ISBN Number0162-8828
KeywordsAutomated;Reproducibility of Results;Sensitivity and Specificity;Social Behavior;, Computer-Assisted;Imaging, Markov model;bee dance;behavior analysis;shape model;social insect;three-tier hierarchical motion model;tracking;video sequence;waggle dance;Markov processes;image sequences;optical tracking;video signal processing;zoology;Algorithms;Animal Communication;, Three-Dimensional;Information Storage and Retrieval;Male;Pattern Recognition
Abstract

Behavior analysis of social insects has garnered impetus in recent years and has led to some advances in fields like control systems and flight navigation. Manual labeling of insect motions required for analyzing the behaviors of insects requires significant investment of time and effort. In this paper, we propose certain general principles that help in simultaneous automatic tracking and behavior analysis, with applications in tracking bees and recognizing specific behaviors that they exhibit. The state space for tracking is defined using the position, orientation, and current behavior of the insect being tracked. The position and the orientation are parameterized using a shape model, whereas the behavior is explicitly modeled using a three-tier hierarchical motion model. The first tier (dynamics) models the local motions exhibited, and the models built in this tier act as a vocabulary for behavior modeling. The second tier is a Markov motion model built on top of the local motion vocabulary, which serves as the behavior model. The third tier of the hierarchy models the switching between behaviors, and this is also modeled as a Markov model. We address issues in learning the three-tier behavioral model, in discriminating between models, and in detecting and modeling abnormal behaviors. Another important aspect of this work is that it leads to joint tracking and behavior analysis instead of the traditional "track-and-then-recognize" approach. We apply these principles for tracking bees in a hive while they are executing the waggle dance and the round dance.

DOI10.1109/TPAMI.2007.70707