Dataset Reader for NeRF synthetic dataset

[MatteoMarengo]

Hi I wish to know why in the dataset_readers.py and especially in readCamerasfromTrasnform the loading of the camera is done like that:

        matrix = np.linalg.inv(np.array(frame["transform_matrix"]))
        R = -np.transpose(matrix[:3, :3])
        R[:, 0] = -R[:, 0]
        T = -matrix[:3, 3]

Indeed in vanilla 3DGS there is an adaptation for COLMAP dataset they use that:

        # NeRF 'transform_matrix' is a camera-to-world transform
        c2w = np.array(frame["transform_matrix"])
        # change from OpenGL/Blender camera axes (Y up, Z back) to COLMAP (Y down, Z forward)
        c2w[:3, 1:3] *= -1

        # get the world-to-camera transform and set R, T
        w2c = np.linalg.inv(c2w)
        R = np.transpose(w2c[:3,:3])  # R is stored transposed due to 'glm' in CUDA code
        T = w2c[:3, 3]

[ingra14m]

Hi, thanks for your interest.

This issue stems from the oldest version of 3D-GS. They solved this problem in this commit.