GitHub - manzinge/Deepstream-Python-YoloV3-Tiny: Sample Application to run YoloV3-Tiny Model with the Deepstream Python Binding

manzinge / Deepstream-Python-YoloV3-Tiny Public

Notifications You must be signed in to change notification settings
Fork 1
Star 5

Sample Application to run YoloV3-Tiny Model with the Deepstream Python Binding

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
.vscode		.vscode
Streams		Streams
__pycache__		__pycache__
nvdsinfer_custom_impl_Yolo		nvdsinfer_custom_impl_Yolo
.gitignore		.gitignore
LICENSE		LICENSE
README		README
deepstream_test_1 live.py		deepstream_test_1 live.py
deepstream_test_video.py		deepstream_test_video.py
dstest1_pgie_config-yolo.txt		dstest1_pgie_config-yolo.txt
labels.txt		labels.txt
mp4_to_h264		mp4_to_h264
plotPicklesArrMat.py		plotPicklesArrMat.py
plotPicklesArrSea.py		plotPicklesArrSea.py
testPlot.py		testPlot.py

Repository files navigation

Prequisites:
- DeepStreamSDK 5.0
- Jetpack 4.4
- Python 3.6
- Gst-python

To run the test app:
  $ python3 deepstream_test_1.py <h264_elementary_stream>

Q: What if I want to use more than 1 Class
A: Modify the PGIE_CLASS Ids in the deepstream_test_1.py, the object and the number
   of classes in the config (dstest1_pgie_config-yolo.txt)

Q: My Stream is lagging 
A: Set the sink property to 0 which will result in smoother framerate
   but also give you slightly off boxes (on the display)