search

NVIDIA-AI-IOT / face-mask-detection

Face Mask Detection using NVIDIA Transfer Learning Toolkit (TLT) and DeepStream for COVID-19
MIT License
241 stars 94 forks source link

How to infer from jetson nano with a camera using deep stream #20

Closed KOOKDONGHUN closed 3 years ago

KOOKDONGHUN commented 3 years ago

I followed the tlt flow provided by GitHub.

Docker pool nvcr.io/nvidia/tlt-streamanalytics:v2.0_py3

I was trained in the docker container above.

I trained on Tesla V100 SXM2 32G multi-GPU environment.

Of the cells provided by face-mask-detection.ipynb, all cells except tlt-evaluate and tlt-infer ran without error.

I would like to use the face mask detection model provided by GitHub on jetson nano for real-time inference via webcam.

What docker should i use

I ran tlt-convert but it still failed to create the .engine file.

How do I create .engine file?

craston commented 3 years ago

@KOOKDONGHUN Did you solve this? If yes, could you please elaborate on the steps.