The formula for camera translation

[ahnHeejune]

in demo.py

camera_translation = torch.stack([pred_camera[:,1], pred_camera[:,2], 2*constants.FOCAL_LENGTH/(constants.IMG_RES * pred_camera[:,0] +1e-9)],dim=-1)

Can you explain why tz the camera translation in z coordinate is

2*constants.FOCAL_LENGTH / constants.IMG_RES * pred_camera[:,0]

= 2 * f / Resolution * scale_factor