def call(self, inputs):
rpn_bbox_deltas = inputs[0]
rpn_labels = inputs[1]
anchors = self.anchors
#
pre_nms_topn = self.hyper_params["pre_nms_topn"]
post_nms_topn = self.hyper_params["train_nms_topn"] if self.mode == "training" else self.hyper_params[
"test_nms_topn"]
nms_iou_threshold = self.hyper_params["nms_iou_threshold"]
variances = self.hyper_params["variances"]
total_anchors = anchors.shape[0]
batch_size = tf.shape(rpn_bbox_deltas)[0]
rpn_bbox_deltas = tf.reshape(rpn_bbox_deltas, (batch_size, total_anchors, 4))
rpn_labels = tf.reshape(rpn_labels, (batch_size, total_anchors))
#
rpn_bbox_deltas *= variances
rpn_bboxes = bbox_utils.get_bboxes_from_deltas(anchors, rpn_bbox_deltas)
#
_, pre_indices = tf.nn.top_k(rpn_labels, pre_nms_topn)
#
pre_roi_bboxes = tf.gather(rpn_bboxes, pre_indices, batch_dims=1)
pre_roi_labels = tf.gather(rpn_labels, pre_indices, batch_dims=1)
#
pre_roi_bboxes = tf.reshape(pre_roi_bboxes, (batch_size, pre_nms_topn, 1, 4))
pre_roi_labels = tf.reshape(pre_roi_labels, (batch_size, pre_nms_topn, 1))
#
roi_bboxes, _, _, _ = bbox_utils.non_max_suppression(pre_roi_bboxes, pre_roi_labels,
max_output_size_per_class=post_nms_topn,
max_total_size=post_nms_topn,
iou_threshold=nms_iou_threshold)
#
return tf.stop_gradient(roi_bboxes)
def call(self, inputs):
roi_bboxes = inputs[0]
pred_deltas = inputs[1]
pred_label_probs = inputs[2]
batch_size = tf.shape(pred_deltas)[0]
#
pred_deltas = tf.reshape(pred_deltas, (batch_size, -1, self.total_labels, 4))
pred_deltas *= self.variances
#
expanded_roi_bboxes = tf.tile(tf.expand_dims(roi_bboxes, -2), (1, 1, self.total_labels, 1))
pred_bboxes = bbox_utils.get_bboxes_from_deltas(expanded_roi_bboxes, pred_deltas)
#
pred_labels_map = tf.expand_dims(tf.argmax(pred_label_probs, -1), -1)
pred_labels = tf.where(tf.not_equal(pred_labels_map, 0), pred_label_probs, tf.zeros_like(pred_label_probs))
#
final_bboxes, final_scores, final_labels, _ = bbox_utils.non_max_suppression(
pred_bboxes, pred_labels,
max_output_size_per_class=self.max_total_size,
max_total_size=self.max_total_size,
score_threshold=self.score_threshold)
#
return final_bboxes, final_labels, final_scores
for image_data in test_data:
img, _, _ = image_data
pred_bbox_deltas, pred_labels = ssd_model.predict_on_batch(img)
#
pred_bbox_deltas *= hyper_params["variances"]
pred_bboxes = bbox_utils.get_bboxes_from_deltas(prior_boxes,
pred_bbox_deltas)
pred_bboxes = tf.clip_by_value(pred_bboxes, 0, 1)
#
pred_labels = tf.cast(pred_labels, tf.float32)
reshaped_pred_bboxes = tf.reshape(
pred_bboxes, (batch_size, pred_bbox_deltas.shape[1], 1, 4))
# Remove background predictions
pred_labels_map = tf.argmax(pred_labels, 2, output_type=tf.int32)
valid_cond = tf.not_equal(pred_labels_map, bg_id)
#
valid_bboxes = tf.expand_dims(reshaped_pred_bboxes[valid_cond], 0)
valid_labels = tf.expand_dims(pred_labels[valid_cond], 0)
#
nms_bboxes, nmsed_scores, nmsed_classes, valid_detections = bbox_utils.non_max_suppression(
valid_bboxes,
valid_labels,
max_output_size_per_class=10,
max_total_size=200,
score_threshold=0.5)
denormalized_bboxes = bbox_utils.denormalize_bboxes(
nms_bboxes[0], img_size, img_size)
drawing_utils.draw_bboxes_with_labels(img[0], denormalized_bboxes,
nmsed_classes[0], nmsed_scores[0],
labels)