Algorithm for generating multiframe.txt

Algorithm for generating multiframe semantic KITTI
n: the number of the frames that will be combined (eg. n = 4, then 4 frames are being combined)
i: the frame that is being added (eg. i = 000000 and n = 4, then 000005, 000010, 000015 are added to 000000)

# repeat for all the files except for the last few
for i in [000000, 000005, .... , 004540 - 5*(n-1)] # the algorithm should be able to find the last data in the directory (in this example the last data is 004540)
    # open i data
    i_bin = get_data(i)

    # read i pose
    i_pose = get_pose(i)

    for j in [i+5*(1), i+5*(2), .... , i+5*(n-1)]
        # open j data
        j_bin = get_data(j)

        # read j pose
        j_pose = get_pose(j)
        
        
        # converts j to world coordinate
        j_bin = to_world_coordinate(j_bin, j_pose)

        # converts j to i's coordinate
        j_bin = to_lidar_coordinate(j_bin, i_pose)
        

        # filter out voxels that are outside of the i's space
        j_bin = filter(j_bin)

        # add filtered voxels into i space
        i_bin = add(i_bin, j_bin)

    # voxelize and save
    voxelize_save(i_bin)