search

theAIGuysCode / yolov4-deepsort

Object tracking implemented with YOLOv4, DeepSort, and TensorFlow.
GNU General Public License v3.0
1.31k stars 745 forks source link

Freeze at Frame #: 1 #119

Open nbourre opened 2 years ago

nbourre commented 2 years ago

I've tried to run the command

python object_tracker.py --weights ./checkpoints/yolov4-tiny-416 --model yolov4 --video ./data/video/test.mp4 --output ./outputs/tiny.avi --tiny

It takes an eternity to get to the print "Frame #:1" and it seems to freeze at the first frame. Nothing is showing.

Can someone guide me in the right direction?

I have a RTX 3600.

PrakharPrashantSingh commented 2 years ago

Yes there is something wrong with RTX 3000 series.. i have RTX 3070, do one thing, uninstall Tensorflow GPU and Install the new version, it will work!! I.e.- Pip uninstall tensorflow-gpu

Pip install tensorflow-gpu

Done

alvarojsj commented 1 year ago

Any updates on that? I got the same issue and it's really annoying...

powerman987 commented 1 year ago

Any updates on that? I got the same issue and it's really annoying...

I wait it for 30min . and it worked. but the result is not ideal

alvarojsj commented 1 year ago

When I wait for some time the video gets processed, but with no detections...

powerman987 commented 1 year ago

I think the calculation is too large, resulting freeze at Frame #1

  1. First ,I tryed tensorflow-cpu ,if it worked. I can confirm the .pb file is fine.
  2. And I change my tensorflow-gpu version from 2.3.0 to 2.4.0 .It worked.