GitLab

changed the interfaces to these functions slightly.  Instead of taking a double that
determines how we decide if boxes match, they now take a test_box_overlap object.

precision values to be slightly wrong.

instead of mean average precision.

since it's easy to mix up the adjust threshold and epsilon arguments
and these checks will usually catch that.  The checks are also very
low cost.

validation routines set the detection threshold to negative infinity in an
attempt to elicit all the possible detections for the purpose of computing mean
average precision scores.  This works fine when using the scan_image_boxes
object but causes the detection quality to degrade when the scan_image_pyramid
is used since this object performs random subsampling to avoid producing an
excessive number of detections.  The fix is to allow the user to set the
adjust_threshold parameter themselves to whatever value they feel is reasonable
rather than hard coding it to negative infinity.

precision measure.

vector_to_matrix(), pointer_to_column_vector(), and pointer_to_matrix() to
mat() (in addition to adding a few more mat() overloads for certain things).

here and there.

both rectangle and full_object_detection objects.

since they logically modify the state of the object detector now that
we are exposing the state of the underlying image scanner.

routines.