omarabid59 / tensorflowdeepsorttracking Goto Github PK

how about the fps of the algorithm?

assert cap.isOpened() == True, 'Could not open Webcam.'

1.How to Write video output from it?
2.Video is looping continuously,how to stop it?

I have a 'video.avi' file in the main directory, and I'd like to run this tracker on it. It seems like the only change I would need to make would be the initial thread_image = WebcamThread('Webcam Thread',1) line into something that locates and opens my video.

But I tried doing that and it didn't seem to work. Any suggestions?

Hey,
I was just curious, what's the point of threads here? Wouldn't GIL just prevent you from utilizing multi CPU setting?

Don't know from where it got pooped up. It is tracking the object but also giving the above error. Can you help?

The video feed is not closing due to the error.

E:\TensorflowDeepSortTracking-master>python object_tracking.py
Traceback (most recent call last):
File "object_tracking.py", line 6, in
from utilities import helper
File "E:\TensorflowDeepSortTracking-master\utilities\helper.py", line 2, in
from object_detection.utils import visualization_utils as vis_util
ModuleNotFoundError: No module named 'object_detection'

ModuleNotFoundError: No module named 'object_detection'?
How can I solve this problem?

Is it possible to read images from the folder. Images have been taken from the panoramic camera on moving bike. How can be achieved with your code?

Thanks.

Hi, I am trying to implement this tracker for custom trained SSD model with 3 classes. But some reason the detections are not seen. What are the other changes need to me made apart from label.pbtxt file and the frozen_inference_graph file?

when doing 'python object_tracking.py' ,turned out "ImportError:No module named ObjectTracking.deep_sort_tracker"

I have had chance to deeply study this project and the original deep sort project. And I review SORT and deep sort paper. Main difference between SORT and deep SORT is deep sort uses deep features of detected objects. Main deep sort repository has a features of MOT dataset as a npc file. And it loads this features. But when we do that in real time we need a deep network for feature extractor. As far as I review this project, there is no operation for feature extract. I just saw only this line for this operation:
feature = [1] self.feature = np.asarray(feature, dtype=np.float32) in here:

Am i miss something?

How can I implement a custom tensorflow model?