def _parse_eval_data(self, data):
        """Generates images and labels that are usable for model training.

    Args:
      data: a dict of Tensors produced by the decoder.
    Returns:
      images: the image tensor.
      labels: a dict of Tensors that contains labels.
    """

        shape = tf.shape(data['image'])
        image = data['image'] / 255
        boxes = data['groundtruth_boxes']
        width = shape[0]
        height = shape[1]

        image, boxes = yolo_preprocess_ops.fit_preserve_aspect_ratio(
            image, boxes, width=width, height=height, target_dim=self._image_w)
        boxes = yolo_box_ops.yxyx_to_xcycwh(boxes)

        # Find the best anchor for the ground truth labels to maximize the iou
        best_anchors = yolo_preprocess_ops.get_best_anchor(
            boxes, self._anchors, width=self._image_w, height=self._image_h)
        boxes = yolo_preprocess_ops.pad_max_instances(boxes,
                                                      self._max_num_instances,
                                                      0)
        classes = yolo_preprocess_ops.pad_max_instances(
            data['groundtruth_classes'], self._max_num_instances, 0)
        best_anchors = yolo_preprocess_ops.pad_max_instances(
            best_anchors, self._max_num_instances, 0)
        area = yolo_preprocess_ops.pad_max_instances(data['groundtruth_area'],
                                                     self._max_num_instances,
                                                     0)
        is_crowd = yolo_preprocess_ops.pad_max_instances(
            tf.cast(data['groundtruth_is_crowd'], tf.int32),
            self._max_num_instances, 0)

        labels = {
            'source_id': data['source_id'],
            'bbox': tf.cast(boxes, self._dtype),
            'classes': tf.cast(classes, self._dtype),
            'area': tf.cast(area, self._dtype),
            'is_crowd': is_crowd,
            'best_anchors': tf.cast(best_anchors, self._dtype),
            'width': width,
            'height': height,
            'num_detections': tf.shape(data['groundtruth_classes'])[0],
        }

        grid = self._build_grid(labels,
                                self._image_w,
                                batch=False,
                                use_tie_breaker=self._use_tie_breaker)
        labels.update({'grid_form': grid})
        return image, labels
Exemple #2
0
 def test_pad_max_instances(self, input_shape, instances, pad_axis):
     expected_output_shape = input_shape
     expected_output_shape[pad_axis] = instances
     output = preprocess_ops.pad_max_instances(np.ones(input_shape),
                                               instances,
                                               pad_axis=pad_axis)
     self.assertAllEqual(expected_output_shape, tf.shape(output).numpy())