def test_with_descending_sorting(self):
sorted_boxlist = np_box_list_ops.sort_by_field(self.boxlist, 'scores')
expected_boxes = np.array([[14.0, 14.0, 15.0, 15.0], [3.0, 4.0, 6.0, 8.0],
[0.0, 0.0, 20.0, 20.0]],
dtype=float)
self.assertAllClose(expected_boxes, sorted_boxlist.get())
expected_scores = np.array([0.9, 0.5, 0.4], dtype=float)
self.assertAllClose(expected_scores, sorted_boxlist.get_field('scores'))
def test_with_descending_sorting(self):
sorted_boxlist = np_box_list_ops.sort_by_field(self.boxlist, 'scores')
expected_boxes = np.array([[14.0, 14.0, 15.0, 15.0], [3.0, 4.0, 6.0, 8.0],
[0.0, 0.0, 20.0, 20.0]],
dtype=float)
self.assertAllClose(expected_boxes, sorted_boxlist.get())
expected_scores = np.array([0.9, 0.5, 0.4], dtype=float)
self.assertAllClose(expected_scores, sorted_boxlist.get_field('scores'))
def sort_by_field(box_mask_list, field,
order=np_box_list_ops.SortOrder.DESCEND):
"""Sort boxes and associated fields according to a scalar field.
A common use case is reordering the boxes according to descending scores.
Args:
box_mask_list: BoxMaskList holding N boxes.
field: A BoxMaskList field for sorting and reordering the BoxMaskList.
order: (Optional) 'descend' or 'ascend'. Default is descend.
Returns:
sorted_box_mask_list: A sorted BoxMaskList with the field in the specified
order.
"""
return box_list_to_box_mask_list(
np_box_list_ops.sort_by_field(
boxlist=box_mask_list, field=field, order=order))
def _get_overlaps_and_scores_relation_tuples(self, detected_box_tuples,
detected_scores,
groundtruth_box_tuples):
"""Computes overlaps and scores between detected and groundtruth tuples.
Both detections and groundtruth boxes have the same class tuples.
Args:
detected_box_tuples: A numpy array of structures with shape [N,],
representing N tuples, each tuple containing the same number of named
bounding boxes.
Each box is of the format [y_min, x_min, y_max, x_max]
detected_scores: A float numpy array of shape [N,], representing
the confidence scores of the detected N object instances.
groundtruth_box_tuples: A float numpy array of structures with the shape
[M,], representing M tuples, each tuple containing the same number
of named bounding boxes.
Each box is of the format [y_min, x_min, y_max, x_max]
Returns:
result_iou: A float numpy array of size
[num_detected_tuples, num_gt_box_tuples].
scores: The score of the detected boxlist.
"""
result_iou = np.ones(
(detected_box_tuples.shape[0], groundtruth_box_tuples.shape[0]),
dtype=float)
for field in detected_box_tuples.dtype.fields:
detected_boxlist_field = np_box_list.BoxList(
detected_box_tuples[field])
detected_boxlist_field.add_field('scores', detected_scores)
detected_boxlist_field = np_box_list_ops.sort_by_field(
detected_boxlist_field, 'scores')
gt_boxlist_field = np_box_list.BoxList(
groundtruth_box_tuples[field])
iou_field = np_box_list_ops.iou(detected_boxlist_field,
gt_boxlist_field)
result_iou = np.minimum(iou_field, result_iou)
scores = detected_boxlist_field.get_field('scores')
return result_iou, scores
def multi_class_non_max_suppression(boxlist, score_thresh, iou_thresh,
max_output_size):
if not 0 <= iou_thresh <= 1.0:
raise ValueError('thresh must be between 0 and 1')
scores = boxlist.get_field('scores')
classes = boxlist.get_field('classes')
num_boxes = boxlist.num_boxes()
selected_boxes_list = []
for cls in np.unique(classes):
inds = np.where(cls == classes)[0]
subboxlist = gather(boxlist, inds)
boxlist_filt = filter_scores_greater_than(subboxlist, score_thresh)
nms_result = non_max_suppression(boxlist_filt,
max_output_size=max_output_size,
iou_threshold=iou_thresh,
score_threshold=score_thresh)
selected_boxes_list.append(nms_result)
selected_boxes = concatenate(selected_boxes_list)
sorted_boxes = sort_by_field(selected_boxes, 'scores')
return sorted_boxes
def multi_class_non_max_suppression_with_extra_fields(
boxlist: BoxList, score_thresh: float, iou_thresh: float,
max_output_size: int) -> BoxList:
"""
Is the same method as
`object_detection.utils.np_box_list_ops.multi_class_non_max_suppression`
but also preserves all the extra fields of the boxlist after nms
"""
# Modification: omit all the assertions
# End of modification
scores = boxlist.get_field('scores')
num_scores = scores.shape[0]
num_classes = scores.shape[1]
selected_boxes_list = []
for class_idx in range(num_classes):
boxlist_and_class_scores = np_box_list.BoxList(boxlist.get())
class_scores = np.reshape(scores[0:num_scores, class_idx], [-1])
boxlist_and_class_scores.add_field('scores', class_scores)
# Modification: add extra fields to boxlist_and_class_scores
for each_extra_field in boxlist.get_extra_fields():
if (not boxlist_and_class_scores.has_field(each_extra_field)
and each_extra_field != "classes"):
boxlist_and_class_scores.add_field(
each_extra_field, boxlist.get_field(each_extra_field))
# End of modification
boxlist_filt = np_box_list_ops.filter_scores_greater_than(
boxlist_and_class_scores, score_thresh)
nms_result = np_box_list_ops.non_max_suppression(
boxlist_filt,
max_output_size=max_output_size,
iou_threshold=iou_thresh,
score_threshold=score_thresh)
nms_result.add_field(
'classes',
np.zeros_like(nms_result.get_field('scores')) + class_idx)
selected_boxes_list.append(nms_result)
selected_boxes = np_box_list_ops.concatenate(selected_boxes_list)
sorted_boxes = np_box_list_ops.sort_by_field(selected_boxes, 'scores')
return sorted_boxes
def _get_overlaps_and_scores_relation_tuples(
self, detected_box_tuples, detected_scores, groundtruth_box_tuples):
"""Computes overlaps and scores between detected and groundtruth tuples.
Both detections and groundtruth boxes have the same class tuples.
Args:
detected_box_tuples: A numpy array of structures with shape [N,],
representing N tuples, each tuple containing the same number of named
bounding boxes.
Each box is of the format [y_min, x_min, y_max, x_max]
detected_scores: A float numpy array of shape [N,], representing
the confidence scores of the detected N object instances.
groundtruth_box_tuples: A float numpy array of structures with the shape
[M,], representing M tuples, each tuple containing the same number
of named bounding boxes.
Each box is of the format [y_min, x_min, y_max, x_max]
Returns:
result_iou: A float numpy array of size
[num_detected_tuples, num_gt_box_tuples].
scores: The score of the detected boxlist.
"""
result_iou = np.ones(
(detected_box_tuples.shape[0], groundtruth_box_tuples.shape[0]),
dtype=float)
for field in detected_box_tuples.dtype.fields:
detected_boxlist_field = np_box_list.BoxList(detected_box_tuples[field])
detected_boxlist_field.add_field('scores', detected_scores)
detected_boxlist_field = np_box_list_ops.sort_by_field(
detected_boxlist_field, 'scores')
gt_boxlist_field = np_box_list.BoxList(groundtruth_box_tuples[field])
iou_field = np_box_list_ops.iou(detected_boxlist_field, gt_boxlist_field)
result_iou = np.minimum(iou_field, result_iou)
scores = detected_boxlist_field.get_field('scores')
return result_iou, scores
def test_with_invalid_sorting_order(self):
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'scores', 'Descending')
def test_with_invalid_field(self):
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'objectness')
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'labels')
def multi_class_non_max_suppression(box_mask_list, score_thresh, iou_thresh,
max_output_size):
"""Multi-class version of non maximum suppression.
This op greedily selects a subset of detection bounding boxes, pruning
away boxes that have high IOU (intersection over union) overlap (> thresh)
with already selected boxes. It operates independently for each class for
which scores are provided (via the scores field of the input box_list),
pruning boxes with score less than a provided threshold prior to
applying NMS.
Args:
box_mask_list: np_box_mask_list.BoxMaskList holding N boxes. Must contain a
'scores' field representing detection scores. This scores field is a
tensor that can be 1 dimensional (in the case of a single class) or
2-dimensional, in which case we assume that it takes the
shape [num_boxes, num_classes]. We further assume that this rank is known
statically and that scores.shape[1] is also known (i.e., the number of
classes is fixed and known at graph construction time).
score_thresh: scalar threshold for score (low scoring boxes are removed).
iou_thresh: scalar threshold for IOU (boxes that that high IOU overlap
with previously selected boxes are removed).
max_output_size: maximum number of retained boxes per class.
Returns:
a box_mask_list holding M boxes with a rank-1 scores field representing
corresponding scores for each box with scores sorted in decreasing order
and a rank-1 classes field representing a class label for each box.
Raises:
ValueError: if iou_thresh is not in [0, 1] or if input box_mask_list does
not have a valid scores field.
"""
if not 0 <= iou_thresh <= 1.0:
raise ValueError('thresh must be between 0 and 1')
if not isinstance(box_mask_list, np_box_mask_list.BoxMaskList):
raise ValueError('box_mask_list must be a box_mask_list')
if not box_mask_list.has_field('scores'):
raise ValueError('input box_mask_list must have \'scores\' field')
scores = box_mask_list.get_field('scores')
if len(scores.shape) == 1:
scores = np.reshape(scores, [-1, 1])
elif len(scores.shape) == 2:
if scores.shape[1] is None:
raise ValueError('scores field must have statically defined second '
'dimension')
else:
raise ValueError('scores field must be of rank 1 or 2')
num_boxes = box_mask_list.num_boxes()
num_scores = scores.shape[0]
num_classes = scores.shape[1]
if num_boxes != num_scores:
raise ValueError('Incorrect scores field length: actual vs expected.')
selected_boxes_list = []
for class_idx in range(num_classes):
box_mask_list_and_class_scores = np_box_mask_list.BoxMaskList(
box_data=box_mask_list.get(),
mask_data=box_mask_list.get_masks())
class_scores = np.reshape(scores[0:num_scores, class_idx], [-1])
box_mask_list_and_class_scores.add_field('scores', class_scores)
box_mask_list_filt = filter_scores_greater_than(
box_mask_list_and_class_scores, score_thresh)
nms_result = non_max_suppression(
box_mask_list_filt,
max_output_size=max_output_size,
iou_threshold=iou_thresh,
score_threshold=score_thresh)
nms_result.add_field(
'classes',
np.zeros_like(nms_result.get_field('scores')) + class_idx)
selected_boxes_list.append(nms_result)
selected_boxes = np_box_list_ops.concatenate(selected_boxes_list)
sorted_boxes = np_box_list_ops.sort_by_field(selected_boxes, 'scores')
return box_list_to_box_mask_list(boxlist=sorted_boxes)
def test_with_invalid_sorting_order(self):
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'scores', 'Descending')
def test_with_invalid_field(self):
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'objectness')
with self.assertRaises(ValueError):
np_box_list_ops.sort_by_field(self.boxlist, 'labels')
