generated from roboflow/template-python
-
Notifications
You must be signed in to change notification settings - Fork 3.1k
Closed
Labels
questionFurther information is requestedFurther information is requested
Description
Search before asking
- I have searched the Supervision issues and found no similar feature requests.
Question
Hi,
I do not understand the precise meaning of lost_track_buffer in the ByteTrack class. It is only used to initialize max_time_lost, which, however, seems to refer to a number of frames rather than a time (and so is a misnomer, IMHO). Furthermore, that variable is initialized as
self.max_time_lost = int(frame_rate / 30.0 * lost_track_buffer)
which does not make sense to me: why is there a division by 30?
Instead, I would understand either of the following possibilities:
- Remove the division by 30. In that case,
lost_frame_bufferrefers to the time (in seconds) a track is retained. - Initialize
max_time_lostwithint(lost_track_buffer). In that case,lost_track_bufferrefers to the number of frames directly, in agreement with the current docstring.
Thanks for clarifying this, and thanks for your great work!
Additional
No response
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
questionFurther information is requestedFurther information is requested