search

michalfaber / tensorflow_Realtime_Multi-Person_Pose_Estimation

Multi-Person Pose Estimation project for Tensorflow 2.0 with a small and fast model based on MobilenetV3
Other
215 stars 65 forks source link

Pose detection using live cam feed #17

Open gautamw3 opened 4 years ago

gautamw3 commented 4 years ago

Hi @michalfaber Can you please provide an insight of how we can detect co-ordinate of key joint locations using live camera. In the older version of this project you have provided this one but not here. Can you help me with this please?

Thanks

AlexandrineRibeiro commented 3 years ago

Hi @gautamw3 You can adapt the demo_video.py file. Just change the parser using @click.option('--video', default=0, type=int, help='Path to the input video file') instead of @click.option('--video', required=True, help='Path to the input video file')

Then, you can use your computer camera by calling: python demo_video.py --video=0 --output-video=sample.mp4 --create-model-fn=create_openpose_singlenet --input-size=224 --output-resize-factor=8 --paf-idx=2 --heatmap-idx=3

Hope it helps.