search

Shimingyi / MotioNet

A deep neural network that directly reconstructs the motion of a 3D human skeleton from monocular video [ToG 2020]
https://rubbly.cn/publications/motioNet/
BSD 2-Clause "Simplified" License
559 stars 83 forks source link

About the output of OpenPose #4

Closed Nisekoixmy closed 3 years ago

Nisekoixmy commented 3 years ago

HI guys,

Thanks for this great work. I tried your code and found it is good on hm3.6 testset. However, I'm wondering what is the output format of OpenPose, because I'm using Python API of OpenPose. And I know there is a 'datum.PoseKeypoints' of OpenPose as outputs. Could you please provide a guideline or a script about what the outputs of OpenPose should be?

Thank you very much.

Best regards, Nisekoi.

Nisekoixmy commented 3 years ago

seems just need json directly from OpenPose? I will give it a try. Thank you.

Shimingyi commented 3 years ago

Right, just need the json file from OpenPose, without any other process:

./build/examples/openpose/openpose.bin --video [input_video] --write_json [output_folder] --display 0 --render_pose 0
Nisekoixmy commented 3 years ago

Right, just need the json file from OpenPose, without any other process:

./build/examples/openpose/openpose.bin --video [input_video] --write_json [output_folder] --display 0 --render_pose 0

Yeah, I tried it same as you provided. Thanks! Btw, it will be great if you provide a flag to control smoothing pre-processing for 2d outputs in evaluate.py.

Anyway, thank you again! I will close this issue.