1. 首先下载官方提供的baloon数据集

import wget
from zipfile import ZipFile


def progress_bar(current, total, width=80):
    progress = current / total
    bar = '#' * int(progress * width)
    percentage = round(progress * 100, 2)
    print(f'[{bar:<{width}}] {percentage}%')


save_path = 'balloon_dataset.zip'
if True:
    url = 'https://github.com/matterport/Mask_RCNN/releases/download/v2.1/balloon_dataset.zip'
    try:
        wget.download(url, save_path, bar=progress_bar)
    except Exception as e:
        print(f'An error occurred: {e}')

extract_path = 'balloon_dataset'
with ZipFile(save_path, "r") as zip:
    zip.printdir()
    zip.extractall(extract_path)

2. 转换成detectron2的格式

# ----------------------------------------------------------------------------
# 转换成 detectron2 的数据格式
# if your dataset is in COCO format, this cell can be replaced by the following three lines:
# from detectron2.data.datasets import register_coco_instances
# register_coco_instances("my_dataset_train", {}, "json_annotation_train.json", "path/to/image/dir")
# register_coco_instances("my_dataset_val", {}, "json_annotation_val.json", "path/to/image/dir")
import cv2
import os
import json
import numpy as np
import random
# import some common detectron2 utilities
from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.config import get_cfg
from detectron2.utils.visualizer import Visualizer
from detectron2.data import MetadataCatalog, DatasetCatalog
from detectron2.structures import BoxMode


def get_balloon_dicts(img_dir):
    json_file = os.path.join(img_dir, "via_region_data.json")
    with open(json_file) as f:
        imgs_anns = json.load(f)

    dataset_dicts = []
    for idx, v in enumerate(imgs_anns.values()):
        record = {}

        filename = os.path.join(img_dir, v["filename"])
        height, width = cv2.imread(filename).shape[:2]

        record["file_name"] = filename
        record["image_id"] = idx
        record["height"] = height
        record["width"] = width

        annos = v["regions"]
        objs = []
        for _, anno in annos.items():
            assert not anno["region_attributes"]
            anno = anno["shape_attributes"]
            px = anno["all_points_x"]
            py = anno["all_points_y"]
            poly = [(x + 0.5, y + 0.5) for x, y in zip(px, py)]
            poly = [p for x in poly for p in x]

            obj = {
                "bbox": [np.min(px), np.min(py), np.max(px), np.max(py)],
                "bbox_mode": BoxMode.XYXY_ABS,
                "segmentation": [poly],
                "category_id": 0,
            }
            objs.append(obj)
        record["annotations"] = objs
        dataset_dicts.append(record)
    return dataset_dicts

# 由于ballon中所有的图像标注都是存储在一个json文件中，所以需要分开，也可以自己保存成单个的json文件
for d in ["train", "val"]:
    DatasetCatalog.register("balloon_" + d, lambda d=d: get_balloon_dicts("balloon_dataset/balloon/" + d))
    MetadataCatalog.get("balloon_" + d).set(thing_classes=["balloon"])
balloon_metadata = MetadataCatalog.get("balloon_train")

dataset_dicts = get_balloon_dicts("balloon_dataset/balloon/train")
for d in random.sample(dataset_dicts, 3):
    img = cv2.imread(d["file_name"])
    visualizer = Visualizer(img[:, :, ::-1], metadata=balloon_metadata, scale=0.5)
    out = visualizer.draw_dataset_dict(d)
    cv2.imshow('img', out.get_image()[:, :, ::-1])
    cv2.waitKey(0)

# ----------------------------------------------------------------------------

3. 训练

# ----------------------------------------------------------------------------
# 转换成 detectron2 的数据格式
# if your dataset is in COCO format, this cell can be replaced by the following three lines:
# from detectron2.data.datasets import register_coco_instances
# register_coco_instances("my_dataset_train", {}, "json_annotation_train.json", "path/to/image/dir")
# register_coco_instances("my_dataset_val", {}, "json_annotation_val.json", "path/to/image/dir")
import cv2
import os
import json
import numpy as np
import random
# import some common detectron2 utilities
from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.engine import DefaultTrainer
from detectron2.config import get_cfg
from detectron2.utils.visualizer import Visualizer
from detectron2.data import MetadataCatalog, DatasetCatalog
from detectron2.structures import BoxMode


def get_balloon_dicts(img_dir):
    json_file = os.path.join(img_dir, "via_region_data.json")
    with open(json_file) as f:
        imgs_anns = json.load(f)

    dataset_dicts = []
    for idx, v in enumerate(imgs_anns.values()):
        record = {}

        filename = os.path.join(img_dir, v["filename"])
        height, width = cv2.imread(filename).shape[:2]

        record["file_name"] = filename
        record["image_id"] = idx
        record["height"] = height
        record["width"] = width

        annos = v["regions"]
        objs = []
        for _, anno in annos.items():
            assert not anno["region_attributes"]
            anno = anno["shape_attributes"]
            px = anno["all_points_x"]
            py = anno["all_points_y"]
            poly = [(x + 0.5, y + 0.5) for x, y in zip(px, py)]
            poly = [p for x in poly for p in x]

            obj = {
                "bbox": [np.min(px), np.min(py), np.max(px), np.max(py)],
                "bbox_mode": BoxMode.XYXY_ABS,
                "segmentation": [poly],
                "category_id": 0,
            }
            objs.append(obj)
        record["annotations"] = objs
        dataset_dicts.append(record)
    return dataset_dicts

# 由于ballon中所有的图像标注都是存储在一个json文件中，所以需要分开，也可以自己保存成单个的json文件
for d in ["train", "val"]:
    DatasetCatalog.register("balloon_" + d, lambda d=d: get_balloon_dicts("balloon_dataset/balloon/" + d))
    MetadataCatalog.get("balloon_" + d).set(thing_classes=["balloon"])
balloon_metadata = MetadataCatalog.get("balloon_train")

cfg = get_cfg()
cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml"))
cfg.DATASETS.TRAIN = ("balloon_train",)
cfg.DATASETS.TEST = ()
cfg.DATALOADER.NUM_WORKERS = 0
cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")  # Let training initialize from model zoo
cfg.SOLVER.IMS_PER_BATCH = 2  # This is the real "batch size" commonly known to deep learning people
cfg.SOLVER.BASE_LR = 0.00025  # pick a good LR
cfg.SOLVER.MAX_ITER = 300    # 300 iterations seems good enough for this toy dataset; you will need to train longer for a practical dataset
cfg.SOLVER.STEPS = []        # do not decay learning rate
cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 128   # The "RoIHead batch size". 128 is faster, and good enough for this toy dataset (default: 512)
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 1  # only has one class (ballon). (see https://detectron2.readthedocs.io/tutorials/datasets.html#update-the-config-for-new-datasets)
# NOTE: this config means the number of classes, but a few popular unofficial tutorials incorrect uses num_classes+1 here.

os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)
trainer = DefaultTrainer(cfg)
trainer.resume_or_load(resume=False)
trainer.train()

4. 测试与评价

# ----------------------------------------------------------------------------
# 转换成 detectron2 的数据格式
# if your dataset is in COCO format, this cell can be replaced by the following three lines:
# from detectron2.data.datasets import register_coco_instances
# register_coco_instances("my_dataset_train", {}, "json_annotation_train.json", "path/to/image/dir")
# register_coco_instances("my_dataset_val", {}, "json_annotation_val.json", "path/to/image/dir")
import cv2
import os
import json
import numpy as np
import random
# import some common detectron2 utilities
from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.engine import DefaultTrainer
from detectron2.config import get_cfg
from detectron2.utils.visualizer import Visualizer
from detectron2.data import MetadataCatalog, DatasetCatalog
from detectron2.structures import BoxMode
from detectron2.utils.visualizer import ColorMode

from detectron2.evaluation import COCOEvaluator, inference_on_dataset
from detectron2.data import build_detection_test_loader
from multiprocessing import freeze_support

def get_balloon_dicts(img_dir):
    json_file = os.path.join(img_dir, "via_region_data.json")
    with open(json_file) as f:
        imgs_anns = json.load(f)

    dataset_dicts = []
    for idx, v in enumerate(imgs_anns.values()):
        record = {}

        filename = os.path.join(img_dir, v["filename"])
        height, width = cv2.imread(filename).shape[:2]

        record["file_name"] = filename
        record["image_id"] = idx
        record["height"] = height
        record["width"] = width

        annos = v["regions"]
        objs = []
        for _, anno in annos.items():
            assert not anno["region_attributes"]
            anno = anno["shape_attributes"]
            px = anno["all_points_x"]
            py = anno["all_points_y"]
            poly = [(x + 0.5, y + 0.5) for x, y in zip(px, py)]
            poly = [p for x in poly for p in x]

            obj = {
                "bbox": [np.min(px), np.min(py), np.max(px), np.max(py)],
                "bbox_mode": BoxMode.XYXY_ABS,
                "segmentation": [poly],
                "category_id": 0,
            }
            objs.append(obj)
        record["annotations"] = objs
        dataset_dicts.append(record)
    return dataset_dicts


if __name__ == '__main__':
    freeze_support()
    # Inference should use the config with parameters that are used in training
    # cfg now already contains everything we've set previously. We changed it a little bit for inference:
    cfg = get_cfg()
    cfg.DATALOADER.NUM_WORKERS = 0
    cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7  # set a custom testing threshold
    cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml"))
    cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth")  # path to the model we just trained
    cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 128  # The "RoIHead batch size". 128 is faster, and good enough for this toy dataset (default: 512)
    cfg.MODEL.ROI_HEADS.NUM_CLASSES = 1  # only has one class (ballon). (see https://detectron2.readthedocs.io/tutorials/datasets.html#update-the-config-for-new-datasets)

    predictor = DefaultPredictor(cfg)

    # 由于ballon中所有的图像标注都是存储在一个json文件中，所以需要分开，也可以自己保存成单个的json文件
    for d in ["train", "val"]:
        DatasetCatalog.register("balloon_" + d, lambda d=d: get_balloon_dicts("balloon_dataset/balloon/" + d))
        MetadataCatalog.get("balloon_" + d).set(thing_classes=["balloon"])
    balloon_metadata = MetadataCatalog.get("balloon_train")

    dataset_dicts = get_balloon_dicts("balloon_dataset/balloon/val")
    for d in random.sample(dataset_dicts, 10):
        im = cv2.imread(d["file_name"])
        outputs = predictor(
            im)  # format is documented at https://detectron2.readthedocs.io/tutorials/models.html#model-output-format
        v = Visualizer(im[:, :, ::-1],
                       metadata=balloon_metadata,
                       scale=0.5,
                       instance_mode=ColorMode.IMAGE
                       # remove the colors of unsegmented pixels. This option is only available for segmentation models
                       )
        out = v.draw_instance_predictions(outputs["instances"].to("cpu"))
        cv2.imshow('result', out.get_image()[:, :, ::-1])
        # cv2.waitKey(0)

    evaluator = COCOEvaluator("balloon_val", output_dir="./output")
    val_loader = build_detection_test_loader(cfg, "balloon_val")
    print(inference_on_dataset(predictor.model, val_loader, evaluator))