search

DeepTrackAI / DeepTrack2

DeepTrack2
MIT License
162 stars 49 forks source link

Lodestar and MAGIK on a 4D dataset #194

Open LRpz opened 1 year ago

LRpz commented 1 year ago

Hello, Firstly, thank you for this package! I find the fusion of strategies employed by Lodestar and MAGIK to be very intuitive for particle tracking.

Secondly, I have a query regarding the modification of the existing examples to accommodate multiple particle detection with Lodestar, and particle linking through MAGIK on a 4D dataset (XYZT).

Also, how are the particle features (like area, eccentricity, Zernike moments, etc.) combined with their xy(z) coodinates to help MAGIK with the linking across frames?

Many thanks in advance!

JesusPinedaC commented 1 year ago

Hello! Incorporating additional features during MAGIK training to improve linking performance is very simple. We have included an extra example to illustrate how to do it! Please let me know if you have any questions.