Tracking Indistinguishable Translucent Objects over Time Using Weakly Supervised Structured Learning
Abstract
We use weakly supervised structured learning to track and disambiguate the identity of multiple indistinguishable, translucent and deformable objects that can overlap for many frames. For this challenging problem, we propose a novel model which handles occlusions, complex motions and non-rigid deformations by jointly optimizing the flows of multiple latent intensities across frames. These flows are latent variables for which the user cannot directly provide labels. Instead, we leverage a structured learning formulation that uses weak user annotations to find the best hyperparameters of this model. The approach is evaluated on a challenging dataset for the tracking of multiple Drosophila larvae which we make publicly available. Our method tracks multiple larvae in spite of their poor distinguishability and minimizes the number of identity switches during prolonged mutual occlusion.
Cite
Text
Fiaschi et al. "Tracking Indistinguishable Translucent Objects over Time Using Weakly Supervised Structured Learning." Conference on Computer Vision and Pattern Recognition, 2014. doi:10.1109/CVPR.2014.356Markdown
[Fiaschi et al. "Tracking Indistinguishable Translucent Objects over Time Using Weakly Supervised Structured Learning." Conference on Computer Vision and Pattern Recognition, 2014.](https://mlanthology.org/cvpr/2014/fiaschi2014cvpr-tracking/) doi:10.1109/CVPR.2014.356BibTeX
@inproceedings{fiaschi2014cvpr-tracking,
title = {{Tracking Indistinguishable Translucent Objects over Time Using Weakly Supervised Structured Learning}},
author = {Fiaschi, Luca and Diego, Ferran and Gregor, Konstantin and Schiegg, Martin and Koethe, Ullrich and Zlatic, Marta and Hamprecht, Fred A.},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2014},
doi = {10.1109/CVPR.2014.356},
url = {https://mlanthology.org/cvpr/2014/fiaschi2014cvpr-tracking/}
}