Replies: 1 comment
-
Hi @Johan154, Currently we do not use any classifiers for tracking. At this point in time, we solve the tracking problem by using a cost function where we compare the distances between instances in sequential frames. Tracking is unnecessary for single animal projects and will only create more work with no reward. Linking a relative discussion #987. Let us know if this helps and/or if you have more questions. Thanks, |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Currently, I am working on a project where I use SLEAP, within this project, only 1 animal is visible per frame. Does tracking help in this situation to increase the accuracy or does this only work for situations with multiple animals within 1 frame?
Can somebody also explain to me what the idea is behind tracking in general? In the documentation it states that tracking can only be used for predicted frames, but I would suppose that it can also be used for training in order to track a certain key point throughout multiple frames.
Beta Was this translation helpful? Give feedback.
All reactions