如何在tensorflow对象检测API中获取多边界框坐标

import tensorflow as tf
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util

# 导入预训练的模型
model = tf.saved_model.load('<模型路径>')
# 加载标签映射文件
label_map = label_map_util.load_labelmap('<标签映射文件路径>')
categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=<类别数量>, use_display_name=True)
category_index = label_map_util.create_category_index(categories)

# 前向推断
output_dict = model(image_tensor)
# 解析输出字典
num_detections = int(output_dict.pop('num_detections'))
output_dict = {key: value[0, :num_detections].numpy() for key, value in output_dict.items()}
output_dict['num_detections'] = num_detections
output_dict['detection_classes'] = output_dict['detection_classes'].astype(np.int64)
# 可选：可视化结果
vis_util.visualize_boxes_and_labels_on_image_array(
    image_np_with_detections,
    output_dict['detection_boxes'],
    output_dict['detection_classes'],
    output_dict['detection_scores'],
    category_index,
    instance_masks=output_dict.get('detection_masks_reframed', None),
    use_normalized_coordinates=True,
    line_thickness=8)

# 获取多个边界框的坐标
num_boxes = len(output_dict['detection_boxes'])
boxes = []
for i in range(num_boxes):
    ymin, xmin, ymax, xmax = output_dict['detection_boxes'][i]
    boxes.append((xmin, ymin, xmax, ymax))