In current version of Isaac Sim, when capturing the 2D bounding boxes of objects in an environment through the camera sensor, there is a very useful feature that is that the user can get the occlusion ratio for each object.
But, sometimes only outputting occlusion ratio is not enough. For example, when the camera is moving, only part of some objects can be seen when they are across the boundary of the FOV of the camera, even it it not occluded by any other objects, like the clock and sofa shown on the uttermost left side of the attached image. In this case, it would be much better if the Isaac Sim can also compute the truncation ratio for each object.
This feature will be very useful in generating the ground truth of object detection and help people to identify which objects should be ignored.