The primary objective of this thesis is to investigate the possibilities of three existing siamese neural networks: SiamRPN++ [1], Siam R-CNN [2], and SiamMOT [3], for tracking cells in two-dimensional time-lapse sequences and to mutually compare their tracking performance. The textual part of the thesis must include a concise and comprehensive description of the three compared approaches and their quantitative evaluation over the Cell Tracking Challenge bioimage data. The tested implementations of the compared approaches will be complemented with a practical guide that documents their exemplary usage.

References:

• [1] B. Li, W. Wu, Q. Wang, F. Zhang, J. Xing, J. Yan. SiamRPN++: Evolution of siamese visual tracking with very deep networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 4282-4291, 2019.
• [2] P. Voigtlaender, J. Luiten, P. H. S. Torr, B. Leibe. Siam R-CNN: Visual tracking by re-detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 6577-6587, 2020.
• [3] B. Shuai, A. Berneshawi, X. Li, D. Modolo, J. Tighe. SiamMOT: Siamese multi-object tracking. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 12367-12377, 2021.