realtime_segmenation
realtime_segmenation copied to clipboard
Realtime Semantic Segmentation for Mobile Platforms based on Tensorflow's Deeblap Model
realtime_segmenation
Realtime semantic segmentation based on Tensorflow's DeepLab Project.
usage
- copy
config.sample.yml
asconfig.yml
, change parameters only inside there - run
demo.py
for the original tensorflow demo - run
test.py
to generate timeline jason files that can be investigated onchrome://tracing/
- run
run.py
for realtime segmentation using openCV
current performance

HELP NEEDED
Please tell me if you are able to run run.py
on your nvidia jetson tx2 and if you get correct mask results!
I face strange results depending on which tensorflow version I use, please tell me your setup!