def build_whole_detection_network(self, input_img_batch, gtboxes_batch):
if self.is_training:
# ensure shape is [M, 5]
gtboxes_batch = tf.reshape(gtboxes_batch, [-1, 5])
gtboxes_batch = tf.cast(gtboxes_batch, tf.float32)
img_shape = tf.shape(input_img_batch)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred, rpn_cls_score, rpn_cls_prob = self.rpn_net(
feature_pyramid)
# 3. generate_anchors
anchors = self.make_anchors(feature_pyramid)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch, anchors],
Tout=[tf.float32, tf.float32, tf.float32])
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
reg_loss = losses.smooth_l1_loss(target_delta, rpn_box_pred,
anchor_states)
self.losses_dict = {
'cls_loss': cls_loss,
'reg_loss': reg_loss * 2
}
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category = postprocess_detctions(
rpn_bbox_pred=rpn_box_pred,
rpn_cls_prob=rpn_cls_prob,
img_shape=img_shape,
anchors=anchors,
is_training=self.is_training)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
if self.is_training:
return boxes, scores, category, self.losses_dict
else:
return boxes, scores, category
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred_list, rpn_cls_score_list, rpn_cls_prob_list = self.rpn_net(
feature_pyramid, 'rpn_net')
# 3. generate_anchors
anchor_list = self.make_anchors(feature_pyramid)
rpn_box_pred = tf.concat(rpn_box_pred_list, axis=0)
rpn_cls_score = tf.concat(rpn_cls_score_list, axis=0)
# rpn_cls_prob = tf.concat(rpn_cls_prob_list, axis=0)
anchors = tf.concat(anchor_list, axis=0)
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.USE_IOU_FACTOR:
reg_loss = losses.iou_smooth_l1_loss(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
self.losses_dict['cls_loss'] = cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['reg_loss'] = reg_loss * cfgs.REG_WEIGHT
with tf.variable_scope('refine_feature_pyramid'):
refine_feature_pyramid = {}
for level in cfgs.LEVEL:
feature_1x5 = slim.conv2d(
inputs=feature_pyramid[level],
num_outputs=256,
kernel_size=[1, 5],
weights_initializer=cfgs.SUBNETS_WEIGHTS_INITIALIZER,
biases_initializer=cfgs.SUBNETS_BIAS_INITIALIZER,
stride=1,
activation_fn=None,
scope='refine_1x5_{}'.format(level))
feature5x1 = slim.conv2d(
inputs=feature_1x5,
num_outputs=256,
kernel_size=[5, 1],
weights_initializer=cfgs.SUBNETS_WEIGHTS_INITIALIZER,
biases_initializer=cfgs.SUBNETS_BIAS_INITIALIZER,
stride=1,
activation_fn=None,
scope='refine_5x1_{}'.format(level))
feature_1x1 = slim.conv2d(
inputs=feature_pyramid[level],
num_outputs=256,
kernel_size=[1, 1],
weights_initializer=cfgs.SUBNETS_WEIGHTS_INITIALIZER,
biases_initializer=cfgs.SUBNETS_BIAS_INITIALIZER,
stride=1,
activation_fn=None,
scope='refine_1x1_{}'.format(level))
refine_feature_pyramid[level] = feature5x1 + feature_1x1
refine_box_pred_list, refine_cls_score_list, refine_cls_prob_list = self.refine_net(
refine_feature_pyramid, 'refine_net')
# refine_box_pred_list, refine_cls_score_list, refine_cls_prob_list = self.refine_net(feature_pyramid, 'refine_net')
refine_box_pred = tf.concat(refine_box_pred_list, axis=0)
refine_cls_score = tf.concat(refine_cls_score_list, axis=0)
refine_cls_prob = tf.concat(refine_cls_prob_list, axis=0)
# refine_boxes = tf.concat(refine_boxes_list, axis=0)
if cfgs.METHOD == 'H':
x_c = (anchors[:, 2] + anchors[:, 0]) / 2
y_c = (anchors[:, 3] + anchors[:, 1]) / 2
h = anchors[:, 2] - anchors[:, 0] + 1
w = anchors[:, 3] - anchors[:, 1] + 1
theta = -90 * tf.ones_like(x_c)
anchors = tf.transpose(tf.stack([x_c, y_c, w, h, theta]))
refine_boxes = bbox_transform.rbbox_transform_inv(boxes=anchors,
deltas=rpn_box_pred)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if not self.is_training:
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category = postprocess_detctions(
refine_bbox_pred=refine_box_pred,
refine_cls_prob=refine_cls_prob,
anchors=refine_boxes,
is_training=self.is_training)
return boxes, scores, category
# 5. build loss
else:
with tf.variable_scope('build_refine_loss'):
refine_labels, refine_target_delta, refine_box_states, refine_target_boxes = tf.py_func(
func=refinebox_target_layer,
inp=[
gtboxes_batch_r, refine_boxes,
cfgs.REFINE_IOU_POSITIVE_THRESHOLD[0],
cfgs.REFINE_IOU_NEGATIVE_THRESHOLD[0], gpu_id
],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
self.add_anchor_img_smry(input_img_batch, refine_boxes,
refine_box_states, 1)
refine_cls_loss = losses.focal_loss(refine_labels,
refine_cls_score,
refine_box_states)
if cfgs.USE_IOU_FACTOR:
refine_reg_loss = losses.iou_smooth_l1_loss(
refine_target_delta, refine_box_pred,
refine_box_states, refine_target_boxes, refine_boxes)
else:
refine_reg_loss = losses.smooth_l1_loss(
refine_target_delta, refine_box_pred,
refine_box_states)
self.losses_dict[
'refine_cls_loss'] = refine_cls_loss * cfgs.CLS_WEIGHT
self.losses_dict[
'refine_reg_loss'] = refine_reg_loss * cfgs.REG_WEIGHT
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category = postprocess_detctions(
refine_bbox_pred=refine_box_pred,
refine_cls_prob=refine_cls_prob,
anchors=refine_boxes,
is_training=self.is_training)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
return boxes, scores, category, self.losses_dict
def refine_stage(self,
input_img_batch,
gtboxes_batch_r,
gt_smooth_label,
box_pred_list,
cls_prob_list,
proposal_list,
angle_cls_list,
feature_pyramid,
gpu_id,
pos_threshold,
neg_threshold,
stage,
proposal_filter=False):
with tf.variable_scope('refine_feature_pyramid{}'.format(stage)):
refine_feature_pyramid = {}
refine_boxes_list = []
refine_boxes_angle_list = []
for box_pred, cls_prob, proposal, angle_prob, stride, level in \
zip(box_pred_list, cls_prob_list, proposal_list, angle_cls_list,
cfgs.ANCHOR_STRIDE, cfgs.LEVEL):
if proposal_filter:
box_pred = tf.reshape(
box_pred, [-1, self.num_anchors_per_location, 5])
proposal = tf.reshape(proposal, [
-1, self.num_anchors_per_location,
5 if self.method == 'R' else 4
])
cls_prob = tf.reshape(
cls_prob,
[-1, self.num_anchors_per_location, cfgs.CLASS_NUM])
cls_max_prob = tf.reduce_max(cls_prob, axis=-1)
box_pred_argmax = tf.cast(
tf.reshape(tf.argmax(cls_max_prob, axis=-1), [-1, 1]),
tf.int32)
indices = tf.cast(
tf.cumsum(tf.ones_like(box_pred_argmax), axis=0),
tf.int32) - tf.constant(1, tf.int32)
indices = tf.concat([indices, box_pred_argmax], axis=-1)
box_pred = tf.reshape(tf.gather_nd(box_pred, indices),
[-1, 5])
proposal = tf.reshape(tf.gather_nd(proposal, indices),
[-1, 5 if self.method == 'R' else 4])
if cfgs.METHOD == 'H':
x_c = (proposal[:, 2] + proposal[:, 0]) / 2
y_c = (proposal[:, 3] + proposal[:, 1]) / 2
h = proposal[:, 2] - proposal[:, 0] + 1
w = proposal[:, 3] - proposal[:, 1] + 1
theta = -90 * tf.ones_like(x_c)
proposal = tf.transpose(
tf.stack([x_c, y_c, w, h, theta]))
else:
box_pred = tf.reshape(box_pred, [-1, 5])
proposal = tf.reshape(proposal, [-1, 5])
bboxes = bbox_transform.rbbox_transform_inv(boxes=proposal,
deltas=box_pred)
if angle_prob is not None:
angle_cls = tf.cast(
tf.argmax(tf.sigmoid(angle_prob), axis=1), tf.float32)
angle_cls = tf.reshape(angle_cls, [
-1,
]) * -1 - 0.5
x, y, w, h, theta = tf.unstack(bboxes, axis=1)
bboxes_angle = tf.transpose(
tf.stack([x, y, w, h, angle_cls]))
refine_boxes_angle_list.append(bboxes_angle)
center_point = bboxes_angle[:, :2] / stride
else:
center_point = bboxes[:, :2] / stride
refine_boxes_list.append(bboxes)
refine_feature_pyramid[level] = self.refine_feature_op(
points=center_point,
feature_map=feature_pyramid[level],
name=level)
refine_box_pred_list, refine_cls_score_list, refine_cls_prob_list, refine_angle_cls_list = self.refine_net(
refine_feature_pyramid, 'refine_net{}'.format(stage))
refine_box_pred = tf.concat(refine_box_pred_list, axis=0)
refine_cls_score = tf.concat(refine_cls_score_list, axis=0)
# refine_cls_prob = tf.concat(refine_cls_prob_list, axis=0)
refine_boxes = tf.concat(refine_boxes_list, axis=0)
refine_angle_cls = tf.concat(refine_angle_cls_list, axis=0)
if self.is_training:
with tf.variable_scope('build_refine_loss{}'.format(stage)):
refine_labels, refine_target_delta, refine_box_states, refine_target_boxes, refine_target_smooth_label = tf.py_func(
func=refinebox_target_layer,
inp=[
gtboxes_batch_r, gt_smooth_label, refine_boxes,
pos_threshold, neg_threshold, gpu_id
],
Tout=[
tf.float32, tf.float32, tf.float32, tf.float32,
tf.float32
])
self.add_anchor_img_smry(input_img_batch, refine_boxes,
refine_box_states, 1)
refine_cls_loss = losses.focal_loss(refine_labels,
refine_cls_score,
refine_box_states)
if False: # cfgs.USE_IOU_FACTOR:
refine_reg_loss = losses.iou_smooth_l1_loss(
refine_target_delta,
refine_box_pred,
refine_box_states,
refine_target_boxes,
refine_boxes,
is_refine=True)
else:
refine_reg_loss = losses.smooth_l1_loss(
refine_target_delta, refine_box_pred,
refine_box_states)
angle_cls_loss = losses.angle_focal_loss(
refine_target_smooth_label, refine_angle_cls,
refine_box_states)
self.losses_dict['refine_cls_loss{}'.format(
stage)] = refine_cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['refine_reg_loss{}'.format(
stage)] = refine_reg_loss * cfgs.REG_WEIGHT
self.losses_dict['angle_cls_loss{}'.format(
stage)] = angle_cls_loss * cfgs.ANGLE_CLS_WEIGHT
return refine_box_pred_list, refine_cls_prob_list, refine_boxes_list, refine_angle_cls_list
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gt_smooth_label,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
gt_smooth_label = tf.reshape(gt_smooth_label,
[-1, cfgs.ANGLE_RANGE])
gt_smooth_label = tf.cast(gt_smooth_label, tf.float32)
img_shape = tf.shape(input_img_batch)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred_list, rpn_cls_score_list, rpn_cls_prob_list = self.rpn_net(
feature_pyramid, 'rpn_net')
# 3. generate_anchors
anchor_list = self.make_anchors(feature_pyramid)
rpn_box_pred = tf.concat(rpn_box_pred_list, axis=0)
rpn_cls_score = tf.concat(rpn_cls_score_list, axis=0)
# rpn_cls_prob = tf.concat(rpn_cls_prob_list, axis=0)
anchors = tf.concat(anchor_list, axis=0)
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.USE_IOU_FACTOR:
reg_loss = losses.iou_smooth_l1_loss(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
self.losses_dict['cls_loss'] = cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['reg_loss'] = reg_loss * cfgs.REG_WEIGHT
box_pred_list, cls_prob_list, proposal_list, angle_cls_list = rpn_box_pred_list, rpn_cls_prob_list, anchor_list, [
None, None, None, None, None
]
all_box_pred_list, all_cls_prob_list, all_proposal_list, all_angle_cls_list = [], [], [], []
for i in range(cfgs.NUM_REFINE_STAGE):
box_pred_list, cls_prob_list, proposal_list, angle_cls_list = self.refine_stage(
input_img_batch,
gtboxes_batch_r,
gt_smooth_label,
box_pred_list,
cls_prob_list,
proposal_list,
angle_cls_list,
feature_pyramid,
gpu_id,
pos_threshold=cfgs.REFINE_IOU_POSITIVE_THRESHOLD[i],
neg_threshold=cfgs.REFINE_IOU_NEGATIVE_THRESHOLD[i],
stage='' if i == 0 else '_stage{}'.format(i + 2),
proposal_filter=True if i == 0 else False)
if not self.is_training:
all_box_pred_list.extend(box_pred_list)
all_cls_prob_list.extend(cls_prob_list)
all_proposal_list.extend(proposal_list)
all_angle_cls_list.extend(angle_cls_list)
else:
all_box_pred_list, all_cls_prob_list, all_proposal_list, all_angle_cls_list = box_pred_list, cls_prob_list, proposal_list, angle_cls_list
with tf.variable_scope('postprocess_detctions'):
box_pred = tf.concat(all_box_pred_list, axis=0)
cls_prob = tf.concat(all_cls_prob_list, axis=0)
proposal = tf.concat(all_proposal_list, axis=0)
angle_cls = tf.concat(all_angle_cls_list, axis=0)
boxes, scores, category, boxes_angle = postprocess_detctions(
refine_bbox_pred=box_pred,
refine_cls_prob=cls_prob,
refine_angle_prob=tf.sigmoid(angle_cls),
anchors=proposal,
is_training=self.is_training)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
boxes_angle = tf.stop_gradient(boxes_angle)
if self.is_training:
return boxes, scores, category, boxes_angle, self.losses_dict
else:
return boxes, scores, category, boxes_angle
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred, rpn_cls_score, rpn_cls_prob = self.rpn_net(
feature_pyramid)
# 3. generate_anchors
anchors = self.make_anchors(feature_pyramid)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if not self.is_training:
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category = postprocess_detctions(
rpn_bbox_pred=rpn_box_pred,
rpn_cls_prob=rpn_cls_prob,
anchors=anchors,
is_training=self.is_training)
return boxes, scores, category
# 5. build loss
else:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors, gpu_id],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.REG_LOSS_MODE == 0:
reg_loss = losses.iou_smooth_l1_loss(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
elif cfgs.REG_LOSS_MODE == 1:
reg_loss = losses.smooth_l1_loss_atan(
target_delta, rpn_box_pred, anchor_states)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
losses_dict = {
'cls_loss': cls_loss * cfgs.CLS_WEIGHT,
'reg_loss': reg_loss * cfgs.REG_WEIGHT
}
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category = postprocess_detctions(
rpn_bbox_pred=rpn_box_pred,
rpn_cls_prob=rpn_cls_prob,
anchors=anchors,
is_training=self.is_training)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
return boxes, scores, category, losses_dict
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gt_encode_label,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
gt_encode_label = tf.reshape(gt_encode_label,
[-1, self.coding_len])
gt_encode_label = tf.cast(gt_encode_label, tf.float32)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred, rpn_cls_score, rpn_cls_prob, rpn_angle_cls = self.rpn_net(
feature_pyramid)
# 3. generate_anchors
anchors = self.make_anchors(feature_pyramid)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes, target_encode_label = tf.py_func(
func=anchor_target_layer,
inp=[
gtboxes_batch_h, gtboxes_batch_r, gt_encode_label,
anchors, gpu_id
],
Tout=[
tf.float32, tf.float32, tf.float32, tf.float32,
tf.float32
])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
reg_loss = losses.smooth_l1_loss(target_delta, rpn_box_pred,
anchor_states)
# angle_cls_loss = losses_dcl.angle_cls_focal_loss(target_encode_label, rpn_angle_cls,
# anchor_states, decimal_weight=None)
angle_cls_loss = losses_dcl.angle_cls_period_focal_loss(
target_encode_label,
rpn_angle_cls,
anchor_states,
target_boxes,
decimal_weight=None)
self.losses_dict = {
'cls_loss': cls_loss * cfgs.CLS_WEIGHT,
'reg_loss': reg_loss * cfgs.REG_WEIGHT,
'angle_cls_loss': angle_cls_loss * cfgs.ANGLE_WEIGHT
}
with tf.variable_scope('postprocess_detctions'):
scores, category, boxes_angle, angle_logits = postprocess_detctions(
rpn_bbox_pred=rpn_box_pred,
rpn_cls_prob=rpn_cls_prob,
rpn_angle_prob=tf.sigmoid(rpn_angle_cls),
rpn_angle_logits=rpn_angle_cls,
anchors=anchors,
is_training=self.is_training)
# boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
boxes_angle = tf.stop_gradient(boxes_angle)
angle_logits = tf.stop_gradient(angle_logits)
if self.is_training:
return scores, category, boxes_angle, angle_logits, self.losses_dict
else:
return scores, category, boxes_angle, angle_logits
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
img_shape = tf.shape(input_img_batch)
# 1. build base network
if cfgs.USE_SUPERVISED_MASK:
feature_pyramid, mask_list, dot_layer_list = self.build_base_network(
input_img_batch)
else:
feature_pyramid = self.build_base_network(input_img_batch)
dot_layer_list = None
mask_list = []
# 2. build rpn
# if cfgs.USE_SUPERVISED_MASK:
# for i, d in enumerate(dot_layer_list):
# feature_pyramid['P{}'.format(i + 3)] *= d
rpn_box_pred_list, rpn_cls_score_list, rpn_cls_prob_list = self.rpn_net(
feature_pyramid, 'rpn_net')
# 3. generate anchors and mask
anchor_list = self.make_anchors(feature_pyramid)
if cfgs.USE_SUPERVISED_MASK:
mask_gt_list = self.generate_mask(mask_list, img_shape,
gtboxes_batch_h, gtboxes_batch_r,
feature_pyramid)
rpn_box_pred = tf.concat(rpn_box_pred_list, axis=0)
rpn_cls_score = tf.concat(rpn_cls_score_list, axis=0)
# rpn_cls_prob = tf.concat(rpn_cls_prob_list, axis=0)
anchors = tf.concat(anchor_list, axis=0)
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.USE_IOU_FACTOR:
reg_loss = losses.iou_smooth_l1_loss_(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
if cfgs.USE_SUPERVISED_MASK:
with tf.variable_scope("supervised_mask_loss"):
mask_loss = 0.0
for i in range(len(mask_list)):
a_mask, a_mask_gt = mask_list[i], mask_gt_list[i]
# b, h, w, c = a_mask.shape
last_dim = 2 if cfgs.BINARY_MASK else cfgs.CLASS_NUM + 1
a_mask = tf.reshape(a_mask, shape=[-1, last_dim])
a_mask_gt = tf.reshape(a_mask_gt, shape=[-1])
a_mask_loss = tf.reduce_mean(
tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=a_mask, labels=a_mask_gt))
mask_loss += a_mask_loss
self.losses_dict[
'mask_loss'] = mask_loss * cfgs.SUPERVISED_MASK_LOSS_WEIGHT / float(
len(mask_list))
self.losses_dict['cls_loss'] = cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['reg_loss'] = reg_loss * cfgs.REG_WEIGHT
box_pred_list, cls_prob_list, proposal_list = rpn_box_pred_list, rpn_cls_prob_list, anchor_list
all_box_pred_list, all_cls_prob_list, all_proposal_list = [], [], []
for i in range(cfgs.NUM_REFINE_STAGE):
box_pred_list, cls_prob_list, proposal_list = self.refine_stage(
input_img_batch,
gtboxes_batch_r,
box_pred_list,
cls_prob_list,
proposal_list,
feature_pyramid,
dot_layer_list,
gpu_id,
pos_threshold=cfgs.REFINE_IOU_POSITIVE_THRESHOLD[i],
neg_threshold=cfgs.REFINE_IOU_NEGATIVE_THRESHOLD[i],
stage='' if i == 0 else '_stage{}'.format(i + 2),
proposal_filter=True if i == 0 else False)
if not self.is_training:
all_box_pred_list.extend(box_pred_list)
all_cls_prob_list.extend(cls_prob_list)
all_proposal_list.extend(proposal_list)
else:
all_box_pred_list, all_cls_prob_list, all_proposal_list = box_pred_list, cls_prob_list, proposal_list
with tf.variable_scope('postprocess_detctions'):
box_pred = tf.concat(all_box_pred_list, axis=0)
cls_prob = tf.concat(all_cls_prob_list, axis=0)
proposal = tf.concat(all_proposal_list, axis=0)
boxes, scores, category = postprocess_detctions(
refine_bbox_pred=box_pred,
refine_cls_prob=cls_prob,
anchors=proposal,
is_training=self.is_training,
gpu_id=gpu_id)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
if self.is_training:
return boxes, scores, category, self.losses_dict
else:
return boxes, scores, category
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gt_encode_label,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
gt_encode_label = tf.reshape(gt_encode_label,
[-1, self.coding_len])
gt_encode_label = tf.cast(gt_encode_label, tf.float32)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred_list, rpn_cls_score_list, rpn_cls_prob_list = self.rpn_net(
feature_pyramid, 'rpn_net')
# 3. generate_anchors
anchor_list = self.make_anchors(feature_pyramid)
rpn_box_pred = tf.concat(rpn_box_pred_list, axis=0)
rpn_cls_score = tf.concat(rpn_cls_score_list, axis=0)
# rpn_cls_prob = tf.concat(rpn_cls_prob_list, axis=0)
anchors = tf.concat(anchor_list, axis=0)
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.USE_IOU_FACTOR:
reg_loss = losses.iou_smooth_l1_loss_(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
self.losses_dict['cls_loss'] = cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['reg_loss'] = reg_loss * cfgs.REG_WEIGHT
with tf.variable_scope('refine_feature_pyramid'):
refine_feature_pyramid = {}
refine_boxes_list = []
for box_pred, cls_prob, anchor, stride, level in \
zip(rpn_box_pred_list, rpn_cls_prob_list, anchor_list,
cfgs.ANCHOR_STRIDE, cfgs.LEVEL):
box_pred = tf.reshape(box_pred,
[-1, self.num_anchors_per_location, 5])
anchor = tf.reshape(anchor, [
-1, self.num_anchors_per_location,
5 if self.method == 'R' else 4
])
cls_prob = tf.reshape(
cls_prob,
[-1, self.num_anchors_per_location, cfgs.CLASS_NUM])
cls_max_prob = tf.reduce_max(cls_prob, axis=-1)
box_pred_argmax = tf.cast(
tf.reshape(tf.argmax(cls_max_prob, axis=-1), [-1, 1]),
tf.int32)
indices = tf.cast(
tf.cumsum(tf.ones_like(box_pred_argmax), axis=0),
tf.int32) - tf.constant(1, tf.int32)
indices = tf.concat([indices, box_pred_argmax], axis=-1)
box_pred_filter = tf.reshape(tf.gather_nd(box_pred, indices),
[-1, 5])
anchor_filter = tf.reshape(tf.gather_nd(
anchor, indices), [-1, 5 if self.method == 'R' else 4])
if cfgs.METHOD == 'H':
x_c = (anchor_filter[:, 2] + anchor_filter[:, 0]) / 2
y_c = (anchor_filter[:, 3] + anchor_filter[:, 1]) / 2
h = anchor_filter[:, 2] - anchor_filter[:, 0] + 1
w = anchor_filter[:, 3] - anchor_filter[:, 1] + 1
theta = -90 * tf.ones_like(x_c)
anchor_filter = tf.transpose(
tf.stack([x_c, y_c, w, h, theta]))
boxes_filter = bbox_transform.rbbox_transform_inv(
boxes=anchor_filter, deltas=box_pred_filter)
refine_boxes_list.append(boxes_filter)
center_point = boxes_filter[:, :2] / stride
refine_feature_pyramid[level] = self.refine_feature_op(
points=center_point,
feature_map=feature_pyramid[level],
name=level)
refine_box_pred_list, refine_cls_score_list, refine_cls_prob_list, refine_angle_cls_list = self.refine_net(
refine_feature_pyramid, 'refine_net')
refine_box_pred = tf.concat(refine_box_pred_list, axis=0)
refine_cls_score = tf.concat(refine_cls_score_list, axis=0)
refine_cls_prob = tf.concat(refine_cls_prob_list, axis=0)
refine_angle_cls = tf.concat(refine_angle_cls_list, axis=0)
refine_boxes = tf.concat(refine_boxes_list, axis=0)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if self.is_training:
with tf.variable_scope('build_refine_loss'):
refine_labels, refine_target_delta, refine_box_states, refine_target_boxes, refine_target_encode_label = tf.py_func(
func=refinebox_target_layer,
inp=[
gtboxes_batch_r, gt_encode_label, refine_boxes,
cfgs.REFINE_IOU_POSITIVE_THRESHOLD[0],
cfgs.REFINE_IOU_NEGATIVE_THRESHOLD[0], gpu_id
],
Tout=[
tf.float32, tf.float32, tf.float32, tf.float32,
tf.float32
])
self.add_anchor_img_smry(input_img_batch, refine_boxes,
refine_box_states, 1)
refine_cls_loss = losses.focal_loss(refine_labels,
refine_cls_score,
refine_box_states)
refine_reg_loss = losses.smooth_l1_loss(
refine_target_delta, refine_box_pred, refine_box_states)
angle_cls_loss = losses_dcl.angle_cls_period_focal_loss(
refine_target_encode_label,
refine_angle_cls,
refine_box_states,
refine_target_boxes,
decimal_weight=cfgs.DATASET_NAME.startswith('DOTA'))
self.losses_dict[
'refine_cls_loss'] = refine_cls_loss * cfgs.CLS_WEIGHT
self.losses_dict[
'refine_reg_loss'] = refine_reg_loss * cfgs.REG_WEIGHT
self.losses_dict[
'angle_cls_loss'] = angle_cls_loss * cfgs.ANGLE_WEIGHT
with tf.variable_scope('postprocess_detctions'):
scores, category, boxes_angle = postprocess_detctions(
refine_bbox_pred=refine_box_pred,
refine_cls_prob=refine_cls_prob,
refine_angle_prob=tf.sigmoid(refine_angle_cls),
refine_boxes=refine_boxes,
is_training=self.is_training)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
boxes_angle = tf.stop_gradient(boxes_angle)
if self.is_training:
return scores, category, boxes_angle, self.losses_dict
else:
return scores, category, boxes_angle
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gt_encode_label,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
gt_encode_label = tf.reshape(gt_encode_label,
[-1, self.coding_len])
gt_encode_label = tf.cast(gt_encode_label, tf.float32)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred_list, rpn_cls_score_list, rpn_cls_prob_list = self.rpn_net(
feature_pyramid, 'rpn_net')
# 3. generate_anchors
anchor_list = self.make_anchors(feature_pyramid)
rpn_box_pred = tf.concat(rpn_box_pred_list, axis=0)
rpn_cls_score = tf.concat(rpn_cls_score_list, axis=0)
# rpn_cls_prob = tf.concat(rpn_cls_prob_list, axis=0)
anchors = tf.concat(anchor_list, axis=0)
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes = tf.py_func(
func=anchor_target_layer,
inp=[gtboxes_batch_h, gtboxes_batch_r, anchors],
Tout=[tf.float32, tf.float32, tf.float32, tf.float32])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.USE_IOU_FACTOR:
reg_loss = losses.iou_smooth_l1_loss_(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
self.losses_dict['cls_loss'] = cls_loss * cfgs.CLS_WEIGHT
self.losses_dict['reg_loss'] = reg_loss * cfgs.REG_WEIGHT
refine_box_pred_list, refine_cls_score_list, refine_cls_prob_list, refine_angle_cls_list = self.refine_net(
feature_pyramid, 'refine_net')
refine_box_pred = tf.concat(refine_box_pred_list, axis=0)
refine_cls_score = tf.concat(refine_cls_score_list, axis=0)
refine_cls_prob = tf.concat(refine_cls_prob_list, axis=0)
refine_angle_cls = tf.concat(refine_angle_cls_list, axis=0)
if cfgs.METHOD == 'H':
x_c = (anchors[:, 2] + anchors[:, 0]) / 2
y_c = (anchors[:, 3] + anchors[:, 1]) / 2
h = anchors[:, 2] - anchors[:, 0] + 1
w = anchors[:, 3] - anchors[:, 1] + 1
theta = -90 * tf.ones_like(x_c)
anchors = tf.transpose(tf.stack([x_c, y_c, w, h, theta]))
# if cfgs.ANGLE_RANGE == 180:
# anchors = tf.py_func(coordinate_present_convert,
# inp=[anchors, -1],
# Tout=[tf.float32])
# anchors = tf.reshape(anchors, [-1, 5])
refine_boxes = bbox_transform.rbbox_transform_inv(boxes=anchors,
deltas=rpn_box_pred)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if self.is_training:
with tf.variable_scope('build_refine_loss'):
refine_labels, refine_target_delta, refine_box_states, refine_target_boxes, refine_target_encode_label = tf.py_func(
func=refinebox_target_layer,
inp=[
gtboxes_batch_r, gt_encode_label, refine_boxes,
cfgs.REFINE_IOU_POSITIVE_THRESHOLD[0],
cfgs.REFINE_IOU_NEGATIVE_THRESHOLD[0], gpu_id
],
Tout=[
tf.float32, tf.float32, tf.float32, tf.float32,
tf.float32
])
self.add_anchor_img_smry(input_img_batch, refine_boxes,
refine_box_states, 1)
refine_cls_loss = losses.focal_loss(refine_labels,
refine_cls_score,
refine_box_states)
refine_reg_loss = losses.smooth_l1_loss(
refine_target_delta, refine_box_pred, refine_box_states)
angle_cls_loss = losses_dcl.angle_cls_period_focal_loss(
refine_target_encode_label,
refine_angle_cls,
refine_box_states,
refine_target_boxes,
decimal_weight=True)
self.losses_dict[
'refine_cls_loss'] = refine_cls_loss * cfgs.CLS_WEIGHT
self.losses_dict[
'refine_reg_loss'] = refine_reg_loss * cfgs.REG_WEIGHT
self.losses_dict[
'angle_cls_loss'] = angle_cls_loss * cfgs.ANGLE_WEIGHT
with tf.variable_scope('postprocess_detctions'):
scores, category, boxes_angle = postprocess_detctions(
refine_bbox_pred=refine_box_pred,
refine_cls_prob=refine_cls_prob,
refine_angle_prob=tf.sigmoid(refine_angle_cls),
refine_boxes=refine_boxes,
is_training=self.is_training)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
boxes_angle = tf.stop_gradient(boxes_angle)
if self.is_training:
return scores, category, boxes_angle, self.losses_dict
else:
return scores, category, boxes_angle
def build_whole_detection_network(self,
input_img_batch,
gtboxes_batch_h,
gtboxes_batch_r,
gthead_quadrant,
gt_smooth_label,
gpu_id=0):
if self.is_training:
gtboxes_batch_h = tf.reshape(gtboxes_batch_h, [-1, 5])
gtboxes_batch_h = tf.cast(gtboxes_batch_h, tf.float32)
gtboxes_batch_r = tf.reshape(gtboxes_batch_r, [-1, 6])
gtboxes_batch_r = tf.cast(gtboxes_batch_r, tf.float32)
gthead_quadrant = tf.reshape(gthead_quadrant, [-1, 1])
gthead_quadrant = tf.cast(gthead_quadrant, tf.int32)
gt_smooth_label = tf.reshape(gt_smooth_label,
[-1, self.angle_range])
gt_smooth_label = tf.cast(gt_smooth_label, tf.float32)
img_shape = tf.shape(input_img_batch)
# 1. build base network
feature_pyramid = self.build_base_network(input_img_batch)
# 2. build rpn
rpn_box_pred, rpn_cls_score, rpn_cls_prob, rpn_head_cls, rpn_angle_cls = self.rpn_net(
feature_pyramid)
# 3. generate_anchors
anchors = self.make_anchors(feature_pyramid)
# 4. postprocess rpn proposals. such as: decode, clip, filter
if self.is_training:
with tf.variable_scope('build_loss'):
labels, target_delta, anchor_states, target_boxes, target_head_quadrant, target_smooth_label = tf.py_func(
func=anchor_target_layer,
inp=[
gtboxes_batch_h, gtboxes_batch_r, gthead_quadrant,
gt_smooth_label, anchors, gpu_id
],
Tout=[
tf.float32, tf.float32, tf.float32, tf.float32,
tf.float32, tf.float32
])
if self.method == 'H':
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 0)
else:
self.add_anchor_img_smry(input_img_batch, anchors,
anchor_states, 1)
cls_loss = losses.focal_loss(labels, rpn_cls_score,
anchor_states)
if cfgs.REG_LOSS_MODE == 0:
reg_loss = losses.iou_smooth_l1_loss(
target_delta, rpn_box_pred, anchor_states,
target_boxes, anchors)
elif cfgs.REG_LOSS_MODE == 1:
reg_loss = losses.smooth_l1_loss_atan(
target_delta, rpn_box_pred, anchor_states)
else:
reg_loss = losses.smooth_l1_loss(target_delta,
rpn_box_pred,
anchor_states)
if cfgs.DATASET_NAME.startswith('DOTA'):
head_cls_loss = losses.head_specific_cls_focal_loss(
target_head_quadrant,
rpn_head_cls,
anchor_states,
labels,
specific_cls=[6, 7, 8, 9, 10, 11])
else:
head_cls_loss = losses.head_focal_loss(
target_head_quadrant, rpn_head_cls, anchor_states)
angle_cls_loss = losses.angle_focal_loss(
target_smooth_label, rpn_angle_cls, anchor_states)
self.losses_dict = {
'cls_loss': cls_loss * cfgs.CLS_WEIGHT,
'reg_loss': reg_loss * cfgs.REG_WEIGHT,
'head_cls_loss': head_cls_loss * cfgs.HEAD_WEIGHT,
'angle_cls_loss': angle_cls_loss * cfgs.ANGLE_WEIGHT
}
with tf.variable_scope('postprocess_detctions'):
boxes, scores, category, boxes_head, boxes_angle = postprocess_detctions(
rpn_bbox_pred=rpn_box_pred,
rpn_cls_prob=rpn_cls_prob,
rpn_angle_prob=tf.sigmoid(rpn_angle_cls),
rpn_head_prob=tf.sigmoid(rpn_head_cls),
anchors=anchors,
is_training=self.is_training)
boxes = tf.stop_gradient(boxes)
scores = tf.stop_gradient(scores)
category = tf.stop_gradient(category)
boxes_head = tf.stop_gradient(boxes_head)
boxes_angle = tf.stop_gradient(boxes_angle)
if self.is_training:
return boxes, scores, category, boxes_head, boxes_angle, self.losses_dict
else:
return boxes, scores, category, boxes_head, boxes_angle
