Visual Computing

University of Konstanz
2014 22nd International Conference on Pattern Recognition

Interactive Framework for Insect Tracking with Active Learning

M. Shen, W. Huang, P. Szyszka, C. Galizia, D. Merhof
Teaser of Interactive Framework for Insect Tracking with Active Learning

Material

Paper (.pdf, 1.2 MB)

Abstract

Extracting motion trajectories of insects is an important prerequisite in many behavioral studies. Despite great efforts to design efficient automatic tracking algorithms, tracking errors are unavoidable. In this paper, we propose general principles that help to minimize the human effort required for accurate multitarget tracking in the form of applications that can track the antennae and mouthparts of a honey bee based on a set of low frame rate videos. This interactive framework estimates which key frames will require user correction, i.e. those that are used for user correction, which are used for 1) incrementally learning an object classifier and 2) data association based tracking. To this framework we apply a standard classification algorithm (i.e. naive Bayesian classification) and an association optimization algorithm (i.e. Hungarian algorithm). The precision of tracking results by our framework on real-world video data is above 98%.

BibTeX

@inproceedings{Shen2014InteractiveFrameworkInsect,
  author     = {M. Shen and W. Huang and P. Szyszka and C. Galizia and D. Merhof},
  booktitle  = {2014 22nd International Conference on Pattern Recognition},
  doi        = {10.1109/ICPR.2014.471},
  issn       = {1051-4651},
  keywords   = {biology computing;image classification;interactive systems;learning (artificial intelligence);sensor fusion;target tracking;video signal processing;association optimization algorithm;classification algorithm;data association;frame rate videos;honey bee antennae tracking;honey bee mouth part tracking;incremental learning;interactive framework;key frame estimation;multitarget tracking;object classifier;Benchmark testing;Insects;Joining processes;Target tracking;Training;Videos;insect tracking;multi-object tracking},
  month      = {aug},
  pages      = {2733-2738},
  title      = {Interactive Framework for Insect Tracking with Active Learning},
  year       = {2014},
  url        = {http://graphics.uni-konstanz.de/publikationen/Shen2014InteractiveFrameworkInsect},
}