Calibridge

Export

points3D.txt is written empty Labels are the image names without folder and extension One line per camera, in the table's order; poses relative to the first camera

    
  
Format conventions

All three use x right, y down, z forward for the camera axes.

COLMAPMetashape XMLopencv_ba.txt
PoseWorld → camera, QW QX QY QZ TX TY TZCamera → chunk 4×4 <transform>; the component (or chunk) transform and its scale map chunk to worldRig → camera, qw qx qy qz tx ty tz; line 1 is the rig root
Pixel gridFirst pixel centre at (0.5, 0.5)Same as COLMAPFirst pixel centre at (0, 0): u0 = cx − 0.5
Focal lengthfx, fyf = fy, b1 = fx − fy, b2 = skewfu = fx, ar = fy/fx, s = skew
Principal pointcx, cycx, cy from the image centreu0, v0
DistortionOPENCV k1 k2 p1 p2; FULL_OPENCV + k3, rational k4–k6; OPENCV_FISHEYE k1–k4Frame: k1–k4, p1–p4 with p1 = OpenCV p2, p2 = OpenCV p1. Fisheye: equidistant k1–k4OpenCV k1 k2 p1 p2 k3, pinhole only
Can't holdSkew, r⁸ (Metashape k4), p3 p4Rational k4–k6Fisheye, r⁸, rational, p3 p4