-
Notifications
You must be signed in to change notification settings - Fork 20
Description
Dear Author
First of all the code was written and commented very nicely. I am currently working on trajectory prediction algorithms. For which we have collected data using a navtech radar and used state of the art detection algorithms to detect vehicles in the radar image. Once detected my next task is to track the vehicles to generate trajectories for each vehicle/target which will then be used for the trajectory prediction algorithm. For tracking the target vehicles I am using the GM PHD filter. I agree that the main task of PHD filter is to remove measurement clutters and not to associate targets across frames. But is there any way I can do this (target association across frames) using this implementation of the GM-PHD filter so that I can generate trajectories for each target?
Best Regards
Saptarshi