search

microsoft / multiview-human-pose-estimation-pytorch

This is an official Pytorch implementation of "Cross View Fusion for 3D Human Pose Estimation, ICCV 2019".
MIT License
541 stars 89 forks source link

Question about camera calibration #9

Closed schmidje closed 4 years ago

schmidje commented 4 years ago

Hi,

It is unclear to me if you somehow provided camera calibration information to your detection network. If yes, how could I provide my own camera calibration information to use your code with my own images acquired from multiple cameras?

Thank you very much!

Best regards

CHUNYUWANG commented 4 years ago

We learn a dedicated model for a particular configuration of cameras. It needs retraining if a new set of cameras is provided. Please see our paper (section 3.2) for more details https://chunyuwang.netlify.com/img/ICCV_Cross_view_camera_ready.pdf

lisa676 commented 4 years ago

@schmidje Hi, did you run this code? and what's the result (2D and 3D) if you have run.

schmidje commented 4 years ago

@lan786 nope, given retraining will require building a training dataset with my camera config which is too much work to do :-/ and not convenient in my opinion. So while I am very interested in this work it is currently not my priority. Sorry!