Problem using ORB-SLAM2 results instead of COLMAP for the data processing step about nerfstudio HOT 1 CLOSED

Solution - needed to apply the coordinate transformation on the predicted poses (OpenCV to OpenGL convention) before writing to the json file. Just defining the "applied_transform" field (which is what I did earlier) does not do this.

Using Nerfstudio's colmap_to_json() code, since we got camera poses (c2w in Nerfstudio's nomenclature) already from ORB SLAM v2, applying this code made things work:

       # Convert from COLMAP's camera coordinate system (OpenCV) to ours (OpenGL)
        c2w[0:3, 1:3] *= -1
        c2w = c2w[np.array([0, 2, 1, 3]), :]
        c2w[2, :] *= -1

from nerfstudio.