Mindkosh Lidar annotation format

Description of the Mindkosh annotation format for cuboid, polyline and segmentation annotation for point clouds.

Versioning

1. v1.1 - This format contains annotations from all frames in a single JSON file.

2. v1.2 - This format has separate JSON files for each frame.

3. v2.0 - This format contains the following additional fields:

   1. version

   2. task name

The following description corresponds to format version 2.0. If you have export files of other formats, you can use conversion scripts - mindkosh1_1tomindkosh2.py and mindkosh1_2tomindkosh2.py (Found in the Python SDK) to convert them into the newer format. All exports created after June 28 2025, will be in format v2.x

Format conventions

We follow the same convention as KITTI, with 1 major change - all our values are specified from the point cloud frame, rather than the camera frame.

Dimensions

1. Length is measured front-to-back.

2. Width is measured side-to-side

3. Height is measured from top-to-bottom

This assumes that the forward facing faces have been properly marked. By default the face is assumed to be in the positive Y direction (when the cuboid is un-rotated).

Rotation

Rotation is measured counter-clockwise from the Y axis (of the point cloud). It is always positive.