def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
#[2000,5] [x1,y1,x2,y2] [2000,1]
rois, rpn_scores = proposal_layer(\
rpn_cls_prob, rpn_bbox_pred, self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
rois, rpn_scores = proposal_layer(rpn_cls_prob, rpn_bbox_pred,
self._im_info, self._mode,
self._feat_stride, self._anchors,
self._num_anchors)
return rois, rpn_scores
def im_detect(self, im):
blobs, im_scales = model_utils._get_blobs(im)
assert len(im_scales) == 1, "Only single-image batch implemented"
im_blob = blobs['data']
blobs['im_info'] = np.array([[im_blob.shape[1], im_blob.shape[2], im_scales[0]]], dtype=np.float32)
anchors, anchor_length = self.generate_anchors(blobs['im_info'])
feed = {
self._images: blobs['data']
}
rpn_cls_prob, rpn_bbox_pred, rpn_feature_map = self.rpn_sess.run([self._rpn_cls_prob, self._rpn_bbox_pred, self._rpn_feature_map], feed)
rois, _ = proposal_layer(rpn_cls_prob, rpn_bbox_pred, blobs['im_info'], 'TEST',
self._feat_stride, anchors, self._num_anchors)
rois = np.reshape(rois, [-1, 5])
# rpn_scores = np.reshape(rpn_scores, [-1, 1])
feed = {
self._rcnn_feature_map: rpn_feature_map,
self._rois: rois
}
scores, bbox_pred = self.rcnn_sess.run([self._cls_prob, self._bbox_pred], feed)
boxes = rois[:, 1:5] / im_scales[0]
scores = np.reshape(scores, [scores.shape[0], -1])
bbox_pred = np.reshape(bbox_pred, [bbox_pred.shape[0], -1])
# Apply bounding-box regression deltas
box_deltas = bbox_pred
pred_boxes = bbox_transform_inv(boxes, box_deltas)
pred_boxes = self._clip_boxes(pred_boxes, im.shape)
return scores, pred_boxes
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
rois, rpn_scores = proposal_layer(\
rpn_cls_prob, rpn_bbox_pred, self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors, self._gt_boxes, self._gt_texts, self._gt_pair)
#print(rois)
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_cls_a_prob, rpn_bbox_pred):
# Training: proposal_layer取前分數前12000的box做nms後再挑前2000個分數高的box
# Testing: proposal_layer取前分數前6000的box做nms後再挑前300個分數高的box
rois, rpn_scores = proposal_layer(\
rpn_cls_prob, rpn_cls_a_prob, rpn_bbox_pred, self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred, name):
with tf.variable_scope(name) as scope:
rois, rpn_scores = proposal_layer(rpn_cls_prob, rpn_bbox_pred,
self._im_info, self._mode,
self._feat_stride, self._anchors,
self._num_anchors)
rois.set_shape([None, 5])
rpn_scores.set_shape([None, 1])
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
rois, rpn_scores = proposal_layer(\
rpn_cls_prob.data.cpu().numpy(), rpn_bbox_pred.data.cpu().numpy(), self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
rois_, rpn_scores_ = proposal_layer_(\
rpn_cls_prob, rpn_bbox_pred, self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
print((torch.Tensor(rois) - rois_.data.cpu()).max())
print((torch.Tensor(rpn_scores) - rpn_scores_.data.cpu()).max())
import utils.timer
rois = rois_.data.cpu().numpy()
rpn_scores = rpn_scores_.data.cpu().numpy()
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred, name):
with tf.variable_scope(name) as scope:
rois, rpn_scores = proposal_layer(
rpn_cls_prob,
rpn_bbox_pred,
self._im_info,
self._mode,
self._feat_stride,
self._anchors,
self._num_anchors
)
rois.set_shape([None, 5])
rpn_scores.set_shape([None, 1])
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
rois, rpn_scores = proposal_layer(\
rpn_cls_prob.data.cpu().numpy(), rpn_bbox_pred.data.cpu().numpy(), self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
return rois, rpn_scores
def _proposal_layer(self, rpn_cls_prob, rpn_bbox_pred):
rois, rpn_scores = proposal_layer(\
rpn_cls_prob, rpn_bbox_pred, self._im_info, self._mode,
self._feat_stride, self._anchors, self._num_anchors)
return rois, rpn_scores
