Closed manerotoni closed 9 months ago
Here I show the btrack results for 20 frames. In red are some examples of what I mean (big jumps in distance)
this is the trackmate results. There are many broken tracks as cells are lost but no big jumps
I may start to answer my own question.
tracker.max_search_radius
limits the linkage distance
Defining the hypotheses I guess should remove splitting, for example use only ["P_FP", "P_init", "P_term", "P_link"]
How the other terms in the hyoptheses config and motionmodel config influences the tracking I still have issues to understand.
May be it would be nice to have different examples config files with some explanations on the results of the tracking. This could help in understand what each of the parameters does and so optimize the parameters for its own purpose.
Overall the plugin is working quite well. Good job!
Hi, in the current tracking problem I would like to track cells in a tissue (2D). There are no divisions/branching. Sometimes a cell gets out of focus and reappear, ideally I would continue to link the track given that the gap is not too big. Also cells do not move much so links above a certain distance should not be considered.
I am more familiar with track mate and there I set max distance and max frame gap and uncheck track segment splitting and there I get what I need
Is there something I can set in the json file to create such a behavior. I tried to change the Hypothesis model parameters, but there was not much happening in the tracks.