How to use this repository for Multi-camera and multi object tracking about yolo_tracking HOT 4 CLOSED

Thank you for the reply, If you don't mind, could you please suggest any alternative repo for the same ID for the object when it passes through multi-camera frames? OR How can I achieve this task by customizing this existing track.py file it would be very helpful :) thank you!!!!

You need a homography to a "global" coordinate space, and then do tracking in that space, and then you can use the reverse of the homography for each camera from the global space. Basically NMS on centroids that have been converted to the global space after all cameras are processed and a custom tracking of ID reference/mapping sets.

from yolo_tracking.

You have two options:

# Multiple streams with batched inference (i.e. batch-size 8 for 8 streams)
source = 'path/to/list.streams'  # *.streams text file with one streaming address per row

or:

# Define a path to a CSV file with images, URLs, videos and directories
source = 'path/to/file.csv'

same ID for the object when it passes through these camera frames?

This we don't support

from yolo_tracking.

Thank you for the reply,
If you don't mind, could you please suggest any alternative repo for the same ID for the object when it passes through multi-camera frames?
OR How can I achieve this task by customizing this existing track.py file
it would be very helpful :)
thank you!!!!

from yolo_tracking.

#594 (comment)

from yolo_tracking.