def match_box_with_gt(self, boxes, iou_threshold):
"""
Args:
boxes: Nx4
Returns:
BoxProposals
"""
if self.is_training:
with tf.name_scope('match_box_with_gt_{}'.format(iou_threshold)):
iou = pairwise_iou(
boxes, self.gt_boxes) # NxM # gt_boxes from datasets
max_iou_per_box = tf.reduce_max(
iou, axis=1) # N # contain many zeros
best_iou_ind = tf.argmax(
iou, axis=1
) # N # if all value is zeros,return the first indices
labels_per_box = tf.gather(self.gt_labels, best_iou_ind)
fg_mask = max_iou_per_box >= iou_threshold
fg_inds_wrt_gt = tf.boolean_mask(
best_iou_ind, fg_mask) # fg_boxes but not gt_boxes
labels_per_box = tf.stop_gradient(labels_per_box *
tf.to_int64(fg_mask))
return BoxProposals(boxes, labels_per_box, fg_inds_wrt_gt,
self.gt_boxes, self.gt_labels)
else:
return BoxProposals(boxes)
def match_box_with_gt(self, boxes, iou_threshold):
"""
Args:
boxes: Nx4
Returns:
BoxProposals
"""
if self.is_training:
with tf.name_scope('match_box_with_gt_{}'.format(iou_threshold)):
iou = pairwise_iou(boxes, self.gt_boxes) # NxM
max_iou_per_box = tf.reduce_max(iou, axis=1) # N
best_iou_ind = tf.argmax(iou, axis=1) # N
labels_per_box = tf.gather(self.gt_labels, best_iou_ind)
fg_mask = max_iou_per_box >= iou_threshold
fg_inds_wrt_gt = tf.boolean_mask(best_iou_ind, fg_mask)
labels_per_box = tf.stop_gradient(labels_per_box *
tf.to_int64(fg_mask))
return BoxProposals(boxes, labels_per_box, fg_inds_wrt_gt,
self.gt_boxes, self.gt_labels)
else:
return BoxProposals(boxes)
def sample_fast_rcnn_targets(boxes, gt_boxes, gt_labels):
"""
Sample some ROIs from all proposals for training.
#fg is guaranteed to be > 0, because grount truth boxes are added as RoIs.
Args:
boxes: nx4 region proposals, floatbox
gt_boxes: mx4, floatbox ,from datasets
gt_labels: m, int32 ,from datasets
Returns:
A BoxProposals instance.
sampled_boxes: tx4 floatbox, the rois
sampled_labels: t int64 labels, in [0, #class). Positive means foreground.
fg_inds_wrt_gt: #fg indices, each in range [0, m-1].
It contains the matching GT of each foreground roi.
"""
iou = pairwise_iou(boxes, gt_boxes) # nxm
proposal_metrics(iou)
# add ground truth as proposals as well
boxes = tf.concat([boxes, gt_boxes], axis=0) # (n+m) x 4
iou = tf.concat([iou, tf.eye(tf.shape(gt_boxes)[0])],
axis=0) # (n x m + m x m) # OK
# #proposal=n+m from now on
def sample_fg_bg(iou):
fg_mask = tf.reduce_max(iou, axis=1) >= cfg.FRCNN.FG_THRESH
fg_inds = tf.reshape(tf.where(fg_mask),
[-1]) # 2-D call mask,1-D call indices
num_fg = tf.minimum(int(cfg.FRCNN.BATCH_PER_IM * cfg.FRCNN.FG_RATIO),
tf.size(fg_inds),
name='num_fg')
fg_inds = tf.random_shuffle(fg_inds)[:num_fg]
bg_inds = tf.reshape(tf.where(tf.logical_not(fg_mask)), [-1])
num_bg = tf.minimum(cfg.FRCNN.BATCH_PER_IM - num_fg,
tf.size(bg_inds),
name='num_bg')
bg_inds = tf.random_shuffle(bg_inds)[:num_bg]
add_moving_summary(num_fg, num_bg) # ??
return fg_inds, bg_inds # len_fg + len_bg = m + n
fg_inds, bg_inds = sample_fg_bg(iou)
# fg,bg indices w.r.t proposals
best_iou_ind = tf.argmax(
iou, axis=1) # #proposal, each in 0~m-1 because after shuffle # OK
fg_inds_wrt_gt = tf.gather(best_iou_ind, fg_inds) # best_fg_indices m
all_indices = tf.concat([fg_inds, bg_inds],
axis=0) # indices w.r.t all n+m proposal boxes
ret_boxes = tf.gather(boxes, all_indices)
ret_labels = tf.concat([
tf.gather(gt_labels, fg_inds_wrt_gt),
tf.zeros_like(bg_inds, dtype=tf.int64)
],
axis=0) # OK
# stop the gradient -- they are meant to be training targets
return BoxProposals(
tf.stop_gradient(ret_boxes, name='sampled_proposal_boxes'),
tf.stop_gradient(ret_labels, name='sampled_labels'),
tf.stop_gradient(fg_inds_wrt_gt), gt_boxes, gt_labels)