Function centers_in_boxes

monai/data/box_utils.py:649–677

Checks which center points are within boxes Args: boxes: bounding boxes, Nx4 or Nx6 torch tensor or ndarray. The box mode is assumed to be ``StandardMode``. centers: center points, Nx2 or Nx3 torch tensor or ndarray. eps: minimum distance to border of boxes. Re

(centers: NdarrayOrTensor, boxes: NdarrayOrTensor, eps: float = 0.01)

def centers_in_boxes(centers: NdarrayOrTensor, boxes: NdarrayOrTensor, eps: float = 0.01) -> NdarrayOrTensor:
    """
    Checks which center points are within boxes

    Args:
        boxes: bounding boxes, Nx4 or Nx6 torch tensor or ndarray. The box mode is assumed to be ``StandardMode``.
        centers: center points, Nx2 or Nx3 torch tensor or ndarray.
        eps: minimum distance to border of boxes.

    Returns:
        boolean array indicating which center points are within the boxes, sized (N,).

    Reference:
        https://github.com/MIC-DKFZ/nnDetection/blob/main/nndet/core/boxes/ops.py

    """
    spatial_dims = get_spatial_dims(boxes=boxes)

    # compute relative position of centers compared to borders
    # should be non-negative if centers are within boxes
    center_to_border = [centers[:, axis] - boxes[:, axis] for axis in range(spatial_dims)] + [
        boxes[:, axis + spatial_dims] - centers[:, axis] for axis in range(spatial_dims)
    ]

    if isinstance(boxes, np.ndarray):
        min_center_to_border: np.ndarray = np.stack(center_to_border, axis=1).min(axis=1)
        return min_center_to_border > eps  # array[bool]

    return torch.stack(center_to_border, dim=1).to(COMPUTE_DTYPE).min(dim=1)[0] > eps  # type: ignore


def boxes_center_distance(