Shape Normalizing and Tracking Dancing Worms
- First Online:
- Cite this paper as:
- Sansone C., Pucher D., Artner N.M., Kropatsch W.G., Saggese A., Vento M. (2016) Shape Normalizing and Tracking Dancing Worms. In: Robles-Kelly A., Loog M., Biggio B., Escolano F., Wilson R. (eds) Structural, Syntactic, and Statistical Pattern Recognition. S+SSPR 2016. Lecture Notes in Computer Science, vol 10029. Springer, Cham
During spawning, the marine worms Platynereis dumerilii exhibit certain swimming behaviors, which are described as nuptial dance. To address the hypothesis that characteristic male and female spawning behaviors are required for successful spawning and fertilization, we propose a 2D tracking approach enabling the extraction of spatio-temporal data to quantify gender-specific behaviors. One of the main issues is the complex interaction between the worms leading to collisions, occlusions, and interruptions of their continuous trajectories. To maintain the individual identities under these challenging interactions a combined tracking and re-identification approach is proposed. The re-identification is based on a set of features, which take into account position, shape and appearance of the worms. These features include the normalized shape of a worm, which is computed using a novel approach based on its distance transform and skeleton.