了解 Sagemaker 对象检测预测的输出

Question

我需要帮助来理解 Amazon Sagemaker 对象检测算法的输出。

这是我的基本目标：识别乒乓球何时在比赛中并在图像帧中标记它的位置。

来自视频源的示例图像：

到目前为止的步骤： 1. 我从一场乒乓球比赛中拍摄了 n 个视频帧。

我用RectLabel手写了乒乓球的位置。
使用 RectLabel，我将这些标签转换为 JSON 文件。此处示例：

{"images":[
    {"id":1,"file_name":"thumb0462.png","width":0,"height":0},
    {"id":2,"file_name":"thumb0463.png","width":0,"height":0},
    {"id":3,"file_name":"thumb0464.png","width":0,"height":0},
    ...
    {"id":4582,"file_name":"thumb6492.png","width":0,"height":0}],
"annotations":[
    {"area":198,"iscrowd":0,"id":1,"image_id":5,"category_id":1,"segmentation":[[59,152,76,152,76,142,59,142]],"bbox":[59,142,18,11]},
    {"area":221,"iscrowd":0,"id":2,"image_id":6,"category_id":1,"segmentation":[[83,155,99,155,99,143,83,143]],"bbox":[83,143,17,13]},
    {"area":399,"iscrowd":0,"id":3,"image_id":8,"category_id":1,"segmentation":[[118,144,136,144,136,124,118,124]],"bbox":[118,124,19,21]},
    {"area":361,"iscrowd":0,"id":4,"image_id":9,"category_id":1,"segmentation":[[132,123,150,123,150,105,132,105]],"bbox":[132,105,19,19]},
    ...
"categories":[{"name":"pp_ball","id":1}]
}

我使用一个函数将注释分成训练和验证文件夹，正如 SageMaker 的输入通道所期望的那样。

file_name = './pp-ball-annotations.json'
with open(file_name) as f:
    js = json.load(f)
    images = js['images']
    categories = js['categories']
    annotations = js['annotations']
    for i in images:
        jsonFile = i['file_name']
        jsonFile = jsonFile.split('.')[0] + '.json'

        line = {}
        line['file'] = i['file_name']
        line['image_size'] = [{
            'width': int(i['width']),
            'height': int(i['height']),
            'depth': 3
        }]
        line['annotations'] = []
        line['categories'] = []
        for j in annotations:
            if j['image_id'] == i['id'] and len(j['bbox']) > 0:
                line['annotations'].append({
                    'class_id': int(j['category_id']),
                    'top': int(j['bbox'][1]),
                    'left': int(j['bbox'][0]),
                    'width': int(j['bbox'][2]),
                    'height': int(j['bbox'][3])
                })
                class_name = ''
                for k in categories:
                    if int(j['category_id']) == k['id']:
                        class_name = str(k['name'])
                assert class_name is not ''
                line['categories'].append({
                    'class_id': int(j['category_id']),
                    'name': class_name
                })
        if line['annotations']:
            with open(os.path.join('generated', jsonFile), 'w') as p:
                json.dump(line, p)

jsons = os.listdir('generated')
print ('There are {} images that have annotation files'.format(len(jsons)))

我按照 SageMaker 的要求将文件移动到具有四个通道（文件夹）的 Amazon S3 存储桶中：/train、/validation、/train_annotation 和 /validation_annotation。

num_annotated_files = len(jsons)
train_split_pct = 0.70
num_train_jsons = int(num_annotated_files * train_split_pct)
random.shuffle(jsons) # randomize/shuffle the JSONs to reduce reliance on *sequenced* frames
train_jsons = jsons[:num_train_jsons]
val_jsons = jsons[num_train_jsons:]

#Moving training files to the training folders
for i in train_jsons:
    image_file = './images/'+i.split('.')[0]+'.png'
    shutil.move(image_file, './train/')
    shutil.move('./generated/'+i, './train_annotation/')

#Moving validation files to the validation folders
for i in val_jsons:
    image_file = './images/'+i.split('.')[0]+'.png'
    shutil.move(image_file, './validation/')
    shutil.move('./generated/'+i, './validation_annotation/')


### Upload to S3
import sagemaker
from sagemaker import get_execution_role

role = sagemaker.get_execution_role()
sess = sagemaker.Session()

from sagemaker.amazon.amazon_estimator import get_image_uri
training_image = get_image_uri(sess.boto_region_name, 'object-detection', repo_version="latest")

bucket = 'pp-balls-object-detection' # custom bucket name.
# bucket = sess.default_bucket()
prefix = 'rect-label-test'

train_channel = prefix + '/train'
validation_channel = prefix + '/validation'
train_annotation_channel = prefix + '/train_annotation'
validation_annotation_channel = prefix + '/validation_annotation'

sess.upload_data(path='train', bucket=bucket, key_prefix=train_channel)
sess.upload_data(path='validation', bucket=bucket, key_prefix=validation_channel)
sess.upload_data(path='train_annotation', bucket=bucket, key_prefix=train_annotation_channel)
sess.upload_data(path='validation_annotation', bucket=bucket, key_prefix=validation_annotation_channel)

s3_train_data = 's3://{}/{}'.format(bucket, train_channel)
s3_validation_data = 's3://{}/{}'.format(bucket, validation_channel)
s3_train_annotation = 's3://{}/{}'.format(bucket, train_annotation_channel)
s3_validation_annotation = 's3://{}/{}'.format(bucket, validation_annotation_channel)

创建了一个具有特定超参数的 SageMaker 对象检测器。我注意到这些超参数 'unusual' 给定了我见过的其他示例：num_classes = 1、use_pretrained_model=0 和 image_shape = 438。

s3_output_location = 's3://{}/{}/output'.format(bucket, prefix)

od_model = sagemaker.estimator.Estimator(training_image,
                                         role,
                                         train_instance_count=1,
                                         train_instance_type='ml.p3.2xlarge',
                                         train_volume_size = 50,
                                         train_max_run = 360000,
                                         input_mode = 'File',
                                         output_path=s3_output_location,
                                         sagemaker_session=sess)

od_model.set_hyperparameters(base_network='resnet-50',
                             use_pretrained_model=0,
                             num_classes=1,
                             mini_batch_size=15,
                             epochs=30,
                             learning_rate=0.001,
                             lr_scheduler_step='10',
                             lr_scheduler_factor=0.1,
                             optimizer='sgd',
                             momentum=0.9,
                             weight_decay=0.0005,
                             overlap_threshold=0.5,
                             nms_threshold=0.45,
                             image_shape=438,
                             label_width=600,
                             num_training_samples=num_train_jsons)

我为对象检测器设置了 train/validate 位置，调用了 .fit 函数，并将模型部署到端点：

train_data = sagemaker.session.s3_input(s3_train_data, distribution='FullyReplicated',
                        content_type='image/png', s3_data_type='S3Prefix')
validation_data = sagemaker.session.s3_input(s3_validation_data, distribution='FullyReplicated',
                             content_type='image/png', s3_data_type='S3Prefix')
train_annotation = sagemaker.session.s3_input(s3_train_annotation, distribution='FullyReplicated',
                             content_type='image/png', s3_data_type='S3Prefix')
validation_annotation = sagemaker.session.s3_input(s3_validation_annotation, distribution='FullyReplicated',
                             content_type='image/png', s3_data_type='S3Prefix')

data_channels = {'train': train_data, 'validation': validation_data,
                 'train_annotation': train_annotation, 'validation_annotation':validation_annotation}

od_model.fit(inputs=data_channels, logs=True)

object_detector = od_model.deploy(initial_instance_count = 1,
                             instance_type = 'ml.m4.xlarge')

我通过以字节为单位传递一个 PNG 文件来调用端点：

file_with_path = 'test/thumb0695.png'
with open(file_with_path, 'rb') as image:
            f = image.read()
            b = bytearray(f)
            ne = open('n.txt', 'wb')
            ne.write(b)

        results = object_detector.predict(b)
        detections = json.loads(results)
        print(detections)

AWS Sagemaker 文档表示期望输出采用以下格式：

Each row in this .json file contains an array that represents a detected object. Each of these object arrays consists of a list of six numbers. The first number is the predicted class label. The second number is the associated confidence score for the detection. The last four numbers represent the bounding box coordinates [xmin, ymin, xmax, ymax]. These output bounding box corner indices are normalized by the overall image size. Note that this encoding is different than that use by the input .json format. For example, in the first entry of the detection result, 0.3088374733924866 is the left coordinate (x-coordinate of upper-left corner) of the bounding box as a ratio of the overall image width, 0.07030484080314636 is the top coordinate (y-coordinate of upper-left corner) of the bounding box as a ratio of the overall image height, 0.7110607028007507 is the right coordinate (x-coordinate of lower-right corner) of the bounding box as a ratio of the overall image width, and 0.9345266819000244 is the bottom coordinate (y-coordinate of lower-right corner) of the bounding box as a ratio of the overall image height.

我们来看一张测试图：

{"id":9,"file_name":"thumb0470.png","width":438,"height":240}

它有一个带有此边界框 [132,105,19,19] 的球（读作 x-top-left, y-top-left, box-width, box-height）。

鉴于我的对象检测器经过训练可以检测一个 class (num_classes=1)，我希望这张图片的输出是这样的：

{'prediction': [[1.0, 0.71, 0.55, 0.239, 0.629, 0.283]]}

相反，我得到了这个输出：

{'prediction': [[0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 0.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0], [0.0, 1.0, 0.0, 0.0, 1.0, 0.0], [0.0, 1.0, 1.0, 0.0, 1.0, 1.0]]}

现在问题：为什么这个模型给我 400 JSON 个元素，而不是一个？

我目前的假设：这个对象检测模型训练的太弱（很有可能，因为这只是第一遍，图像太少），单发检测器正在识别它认为是 400 个实例的对象"ping pong ball" 在图像中。

但即使我的假设是正确的，为什么输出重复这么多？有 178 个相同的 'predictions' 形式

[0.0, 1.0, 0.0, 0.0, 1.0, 0.0]

如果解释为：

0.0 - class 我没有定义的对象“0”。所以我假设这意味着 "no ball in play"

1.0 - 100% 置信度

0.0 - 作为宽度比率的 xmin 位置 = 0

0.0 - ymin 位置与高度的比率 = 0

1.0 - 作为宽度比率的 xmax 位置 = 240

0.0 - 作为高度比率的 ymax 位置 = 0

坐标 [xmin: 0, ymin: 0, xmax: 240, ymax: 0] 就像在第一个像素上画一条线。

感谢您的帮助！

-------- 根据 Ryo 的回答进行编辑 ------

将类别 ID 重新映射到索引基 0 非常有效。以下是仅 2,000 张标记图像的结果：

这是 Ryo 的有用回答后的代码：

def fixCategoryId(category_id):
    return category_id - 1;

with open(file_name) as f:
    js = json.load(f)
    images = js['images']
    categories = js['categories']
    annotations = js['annotations']
    for i in images:
        jsonFile = i['file_name']
        jsonFile = jsonFile.split('.')[0] + '.json'

        line = {}
        line['file'] = i['file_name']
        line['image_size'] = [{
            'width': int(i['width']),
            'height': int(i['height']),
            'depth': 3
        }]
        line['annotations'] = []
        line['categories'] = []
        for j in annotations:
            if j['image_id'] == i['id'] and len(j['bbox']) > 0:
                line['annotations'].append({
                    'class_id': fixCategoryId(int(j['category_id'])),
                    'top': int(j['bbox'][1]),
                    'left': int(j['bbox'][0]),
                    'width': int(j['bbox'][2]),
                    'height': int(j['bbox'][3])
                })
                class_name = ''
                for k in categories:
                    if int(j['category_id']) == k['id']:
                        class_name = str(k['name'])
                assert class_name is not ''
                line['categories'].append({
                    'class_id': fixCategoryId(int(j['category_id'])),
                    'name': class_name
                })
        if line['annotations']:
            with open(os.path.join('generated', jsonFile), 'w') as p:
                json.dump(line, p)

jsons = os.listdir('generated')
print ('There are {} images that have annotation files'.format(len(jsons)))

Answer 1

发件人：https://docs.aws.amazon.com/sagemaker/latest/dg/algo-object-detection-tech-notes.html

Object detection algorithm identifies and locates all instances of objects in an image from a known collection of object categories

这解释了为什么您从 predict 返回响应数组中的 400 项。

这里输出的数据有误。您已将它正确映射到图像的顶部，但它的高度为 0，因此它基本上无处可去。

发件人：https://docs.aws.amazon.com/sagemaker/latest/dg/object-detection.html#object-detection-inputoutput

The "categories" property stores the mapping between the class index and class name. The class indices should be numbered successively and the numbering should start with 0. The "categories" property is optional for the annotation .json file

您的类别数组以 class id 1 开头。

您提供了从标签工具获得的样本 json，但文件 generated 中没有您生成的 json 的样本。查看该输出的示例也会很有帮助。

Answer 2

尽管 COCO JSON 文件中的 'category_id' 从 1 开始，但 Amazon SageMaker JSON 文件中的 'class_id' 从 0 开始。

你的转换代码应该是这样的。

def fixCategoryId(category_id):
    return category_id - 1;

with open(coco_json_path) as f:
    js = json.load(f)
    images = js['images']
    categories = js['categories']
    annotations = js['annotations']
    for i in images:
        jsonFile = i['file_name']
        jsonFile = jsonFile.split('.')[0] + '.json'

        line = {}
        line['file'] = i['file_name']
        line['image_size'] = [{
            'width': int(i['width']),
            'height': int(i['height']),
            'depth': 3
        }]
        line['annotations'] = []
        line['categories'] = []
        for j in annotations:
            if j['image_id'] == i['id'] and len(j['bbox']) > 0:
                line['annotations'].append({
                    'class_id': fixCategoryId(int(j['category_id'])),
                    'top': int(j['bbox'][1]),
                    'left': int(j['bbox'][0]),
                    'width': int(j['bbox'][2]),
                    'height': int(j['bbox'][3])
                })
                class_name = ''
                for k in categories:
                    if int(j['category_id']) == k['id']:
                        class_name = str(k['name'])
                assert class_name is not ''
                line['categories'].append({
                    'class_id': fixCategoryId(int(j['category_id'])),
                    'name': class_name
                })
        if line['annotations']:
            with open(os.path.join(sagemaker_json_path, jsonFile), 'w') as p:
                json.dump(line, p)

In the Amazon SageMaker doc，他们正在使用 get_coco_mapper()。

import json
import logging

def get_coco_mapper():
    original_list = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20,
                    21, 22, 23, 24, 25, 27, 28, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
                    41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60,
                    61, 62, 63, 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80,
                    81, 82, 84, 85, 86, 87, 88, 89, 90]
    iter_counter = 0
    COCO = {}
    for orig in original_list:
        COCO[orig] = iter_counter
        iter_counter += 1
    return COCO

训练完模型后，您必须检查每个损失是否减少。

od_model.fit(inputs=data_channels, logs=True)

[11/04/2019 09:26:46 INFO 140651482974016] #quality_metric: host=algo-1, epoch=499, batch=11 train cross_entropy <loss>=(0.20304460724736212)
[11/04/2019 09:26:46 INFO 140651482974016] #quality_metric: host=algo-1, epoch=499, batch=11 train smooth_l1 <loss>=(0.06970448779799958)

如果您有任何问题，请告诉我们。

了解 Sagemaker 对象检测预测的输出

understanding the output of Sagemaker Object Detection prediction

python

machine-learning

object-detection

computer-vision

amazon-sagemaker