I train the VoxelPose using the Panoptic dataset.
So, I try to implement the trained model using our custom dataset.
However, our five cameras do not have camera calibration information (K, distCoef, R, and t).
We just have 5 videos recorded same time.
We have the 2D pose extracted by each frame in each video.
We have 3D pose from each frame.
Now,
How to make camera calibration in this situation?
- It can be multiview calibration or calibrate each camera seperately.
I am trying Open CV2 in Python like cv2.calibrateCamera
But i cannot get correct calibration.
Tserenpurev Chuluunsaikhan is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.