In target tracking, the dimensions of objects change, especially if they are detected using a LIDAR sensor. Also, the static objects in consecutive frames are not 100% static, their position changes a little bit, due to the point cloud segmentation algorithms (which is somehow expected).

After I associate a tracklet (which maintains the objects previous dimension) and a measurement (that has changed its dimension in the current frame) and perform a Kalman update, a small velocity is induced in the new updated tracklet, even if my object is static (I am considering the reference point of the object and tracking its center).

Is there any solution for not inducing and displaying such a velocity in the updated tracklets?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.