目标检测
目标检测是计算机视觉中的一项任务,旨在检测图像中的实例(例如人、建筑物或汽车)。目标检测模型以图像作为输入,输出检测到的对象的边界框坐标及其对应的标签。一张图像可以包含多个对象,每个对象都有各自的边界框和标签(例如,图像中可能同时包含一辆汽车和一座建筑物),并且每个对象可以出现在图像的不同位置(例如,图像中可能有多辆汽车)。该任务常用于自动驾驶领域,以检测行人、交通标志和交通信号灯等。其他应用包括图像中的对象计数、图像搜索等。
在本指南中,您将学习如何:
- 在 mobile-ui-design 数据集上微调 RF-DETR,以检测移动应用截图中的 UI 元素。
- 使用微调后的模型进行推理。
要查看与此任务兼容的所有架构和检查点,我们建议查阅 任务页面
在开始之前,请确保已安装所有必要的库:
pip install -q datasets transformers accelerate timm trackio torchmetrics pycocotools
您将使用 🤗 Datasets 从 Hugging Face Hub 加载数据集,并使用 🤗 Transformers 训练您的模型。
我们鼓励您与社区分享您的模型。请登录您的 Hugging Face 账户,将其上传到 Hub。 当系统提示时,请输入您的 token 以完成登录:
>>> from huggingface_hub import notebook_login
>>> notebook_login()定义全局常量,即模型名称和图像尺寸。本教程使用 RF-DETR,但您可以选择 Transformers 中的任意目标检测模型。
>>> MODEL_NAME = "Roboflow/rf-detr-medium"加载 mobile-ui-design 数据集
mobile-ui-design 数据集包含带有注释的移动应用截图,用于检测文本、图像、矩形和组等 UI 元素。
首先加载数据集并提取类别标签。该数据集已经包含数据划分。
>>> from datasets import load_dataset
>>> ds = load_dataset("merve/mobile-ui-design")
>>> CATEGORIES = sorted(set(
... cat for split in ds.values() for example in split for cat in example["objects"]["category"]
... ))
>>> label2id = {label: i for i, label in enumerate(CATEGORIES)}
>>> id2label = {i: label for label, i in label2id.items()}
>>> print(f"Categories ({len(CATEGORIES)}): {CATEGORIES}")
Categories (4): ['group', 'image', 'rectangle', 'text']该数据集使用字符串类别名称和 COCO 格式的边界框 (x, y, w, h)。在训练之前,请将类别转换为整数 ID,计算面积,并过滤掉退化的边界框:
>>> def prepare_example(example, idx):
... objects = example["objects"]
... in_bboxes = objects["bbox"]
... categories = objects["category"]
... img_w, img_h = example["width"], example["height"]
... bboxes, cats, areas, ids = [], [], [], []
... for i, (bbox, cat) in enumerate(zip(in_bboxes, categories)):
... x, y, w, h = bbox
... if w <= 0 or h <= 0:
... continue
... x = max(0.0, min(x, img_w))
... y = max(0.0, min(y, img_h))
... w = min(w, img_w - x)
... h = min(h, img_h - y)
... if w <= 0 or h <= 0:
... continue
... bboxes.append([x, y, w, h])
... cats.append(label2id[cat])
... areas.append(w * h)
... ids.append(i)
... return {
... "image_id": idx, "image": example["image"],
... "width": example["width"], "height": example["height"],
... "objects": {"id": ids, "bbox": bboxes, "category": cats, "area": areas},
... }
>>> ds_prepared = ds["train"].map(prepare_example, with_indices=True, remove_columns=ds["train"].column_names)
>>> ds_prepared = ds_prepared.filter(lambda x: len(x["objects"]["bbox"]) > 0)
>>> split = ds_prepared.train_test_split(test_size=0.15, seed=1337)
>>> train_ds = split["train"]
>>> val_ds = split["test"]
>>> print(f"Train: {len(train_ds)}, Validation: {len(val_ds)}")
Train: 6669, Validation: 1177预处理数据
AutoImageProcessor 负责处理图像数据,生成模型可用于训练的 pixel_values、pixel_mask 和 labels。图像处理器会处理缩放、填充和归一化。此外,你还可以选择性地添加随机数据增强(参见下文)以提升泛化能力。
>>> import numpy as np
>>> from functools import partial
>>> from transformers import AutoImageProcessor
>>> image_processor = AutoImageProcessor.from_pretrained(MODEL_NAME)image_processor 期望以 COCO 格式提供标注:{'image_id': int, 'annotations': list[Dict]}。请格式化每个示例的标注,其余部分交由处理器处理:
>>> def format_image_annotations_as_coco(image_id, categories, areas, bboxes):
... annotations = []
... for category, area, bbox in zip(categories, areas, bboxes):
... annotations.append({
... "image_id": image_id,
... "category_id": category,
... "iscrowd": 0,
... "area": area,
... "bbox": list(bbox),
... })
... return {"image_id": image_id, "annotations": annotations}
>>> def transform_batch(examples, image_processor):
... images = []
... annotations = []
... for image_id, image, objects in zip(examples["image_id"], examples["image"], examples["objects"]):
... images.append(np.array(image.convert("RGB")))
... formatted = format_image_annotations_as_coco(
... image_id, objects["category"], objects["area"], objects["bbox"]
... )
... annotations.append(formatted)
... result = image_processor(images=images, annotations=annotations, return_tensors="pt")
... result.pop("pixel_mask", None)
... return result
>>> transform_fn = partial(transform_batch, image_processor=image_processor)
>>> train_ds = train_ds.with_transform(transform_fn)
>>> val_ds = val_ds.with_transform(transform_fn)数据增强
上述转换仅对图像进行缩放和归一化。通常,对 训练 集应用随机数据增强可以提升模型的泛化能力,而验证集应保持无增强,以确保评估结果的确定性。一个常见的选择是 Albumentations,它可以同时对图像及其边界框进行增强。使用 bbox_params 定义一个 pipeline,使边界框与图像保持一致的变换,然后根据增强后的边界框重新计算面积:
>>> import albumentations as A
>>> train_augment = A.Compose(
... [
... A.Perspective(p=0.1),
... A.HorizontalFlip(p=0.5),
... A.RandomBrightnessContrast(p=0.5),
... A.HueSaturationValue(p=0.1),
... ],
... bbox_params=A.BboxParams(format="coco", label_fields=["category"], clip=True, min_area=25),
... )
>>> def augment_and_transform_batch(examples, image_processor, transform):
... images = []
... annotations = []
... for image_id, image, objects in zip(examples["image_id"], examples["image"], examples["objects"]):
... image = np.array(image.convert("RGB"))
... output = transform(image=image, bboxes=objects["bbox"], category=objects["category"])
... images.append(output["image"])
... areas = [w * h for (_, _, w, h) in output["bboxes"]]
... formatted = format_image_annotations_as_coco(
... image_id, output["category"], areas, output["bboxes"]
... )
... annotations.append(formatted)
... result = image_processor(images=images, annotations=annotations, return_tensors="pt")
... result.pop("pixel_mask", None)
... return result仅对训练集应用增强转换,验证集保留原始的 transform_fn:
>>> train_augment_fn = partial(augment_and_transform_batch, image_processor=image_processor, transform=train_augment)
>>> train_ds = train_ds.with_transform(train_augment_fn)
>>> val_ds = val_ds.with_transform(transform_fn)创建一个自定义的 collate_fn 以将图像进行批量处理:
>>> import torch
>>> def collate_fn(batch):
... data = {}
... data["pixel_values"] = torch.stack([x["pixel_values"] for x in batch])
... data["labels"] = [x["labels"] for x in batch]
... if "pixel_mask" in batch[0]:
... data["pixel_mask"] = torch.stack([x["pixel_mask"] for x in batch])
... return data
准备用于计算 mAP 的函数
目标检测模型通常使用一组 COCO 风格的指标 进行评估。我们将使用 torchmetrics 来计算 mAP(平均精度均值)和 mAR(平均召回率均值)指标,并将其封装为 compute_metrics 函数,以便在 Trainer 中用于评估。
用于训练的框的中间格式为 YOLO(归一化),但为了正确处理框的面积,我们将以 Pascal VOC(绝对值)格式计算框的指标。让我们定义一个函数,将边界框转换为 Pascal VOC 格式:
>>> from transformers.image_transforms import center_to_corners_format
>>> def convert_bbox_yolo_to_pascal(boxes, image_size):
... """
... Convert bounding boxes from YOLO format (x_center, y_center, width, height) in range [0, 1]
... to Pascal VOC format (x_min, y_min, x_max, y_max) in absolute coordinates.
... Args:
... boxes (torch.Tensor): Bounding boxes in YOLO format
... image_size (tuple[int, int]): Image size in format (height, width)
... Returns:
... torch.Tensor: Bounding boxes in Pascal VOC format (x_min, y_min, x_max, y_max)
... """
... # convert center to corners format
... boxes = center_to_corners_format(boxes)
... # convert to absolute coordinates
... height, width = image_size
... boxes = boxes * torch.tensor([[width, height, width, height]])
... return boxes然后,在 compute_metrics 函数中,我们从评估循环的结果中收集 predicted 和 target 的边界框、分数和标签,并将其传递给评分函数。
>>> import numpy as np
>>> from dataclasses import dataclass
>>> from torchmetrics.detection.mean_ap import MeanAveragePrecision
>>> @dataclass
>>> class ModelOutput:
... logits: torch.Tensor
... pred_boxes: torch.Tensor
>>> def _get_orig_size(image_target):
... """Robust orig_size extraction - Trainer serialization can truncate to 1 element."""
... orig = np.atleast_1d(np.asarray(image_target["orig_size"])).flatten()
... if len(orig) >= 2:
... return (int(orig[0]), int(orig[1]))
... return (int(orig[0]), int(orig[0]))
>>> @torch.no_grad()
>>> def compute_metrics(evaluation_results, image_processor, threshold=0.0, id2label=None):
... predictions, targets = evaluation_results.predictions, evaluation_results.label_ids
... image_sizes = []
... post_processed_targets = []
... post_processed_predictions = []
...
... for batch in targets:
... batch_sizes = []
... for image_target in batch:
... h, w = _get_orig_size(image_target)
... batch_sizes.append([h, w])
... boxes = torch.tensor(image_target["boxes"])
... boxes = convert_bbox_yolo_to_pascal(boxes, (h, w))
... labels = torch.tensor(image_target["class_labels"])
... post_processed_targets.append({"boxes": boxes, "labels": labels})
... image_sizes.append(torch.tensor(batch_sizes))
...
... for batch, target_sizes in zip(predictions, image_sizes):
... batch_logits, batch_boxes = batch[1], batch[2]
... output = ModelOutput(logits=torch.tensor(batch_logits), pred_boxes=torch.tensor(batch_boxes))
... post_processed_output = image_processor.post_process_object_detection(
... output, threshold=threshold, target_sizes=target_sizes
... )
... post_processed_predictions.extend(post_processed_output)
...
... metric = MeanAveragePrecision(box_format="xyxy", class_metrics=True)
... metric.update(post_processed_predictions, post_processed_targets)
... metrics = metric.compute()
...
... classes = metrics.pop("classes")
... map_per_class = metrics.pop("map_per_class")
... mar_100_per_class = metrics.pop("mar_100_per_class")
... for class_id, class_map, class_mar in zip(classes, map_per_class, mar_100_per_class):
... class_name = id2label[class_id.item()] if id2label is not None else class_id.item()
... metrics[f"map_{class_name}"] = class_map
... metrics[f"mar_100_{class_name}"] = class_mar
...
... metrics = {k: round(v.item(), 4) for k, v in metrics.items()}
... return metrics
>>> eval_compute_metrics_fn = partial(
... compute_metrics, image_processor=image_processor, id2label=id2label, threshold=0.0
... )训练检测模型
在之前的章节中,您已经完成了大部分繁重的工作,现在可以开始训练您的模型了! 该数据集中的图像即使经过调整大小后仍然相当大。这意味着微调该模型至少需要一块 GPU。
训练包括以下步骤:
- 使用与预处理中相同的检查点,通过 AutoModelForObjectDetection 加载模型。
- 在 TrainingArguments 中定义你的训练超参数。
- 将训练参数与模型、数据集、图像处理器和数据整理器一起传递给 Trainer。
- 调用 train() 来微调你的模型。
从用于预处理的同一检查点加载模型时,请记得传入之前从数据集元数据中创建的 label2id 和 id2label 映射。此外,我们指定 ignore_mismatched_sizes=True,以用新的分类头替换现有的分类头。
>>> from transformers import AutoModelForObjectDetection
>>> model = AutoModelForObjectDetection.from_pretrained(
... MODEL_NAME,
... id2label=id2label,
... label2id=label2id,
... ignore_mismatched_sizes=True,
... )在 TrainingArguments 中,使用 output_dir 指定模型的保存位置,然后按需配置超参数。对于 num_train_epochs=5,在 A100 GPU 上训练大约需要 35 分钟,增加训练轮数可获得更好的效果。
重要说明:
- 不要删除未使用的列,因为这会丢弃图像列。如果没有图像列,就无法创建
pixel_values。因此,请将remove_unused_columns设置为False。 - 将
eval_do_concat_batches=False设置为 False 以获得正确的评估结果。由于不同图像的目标框数量不同,如果将批次进行拼接,将无法确定哪些框属于特定图像。
如果您希望通过推送到 Hub 来共享您的模型,请将 push_to_hub 设置为 True(您必须已登录 Hugging Face 才能上传模型)。
>>> from transformers import TrainingArguments
>>> training_args = TrainingArguments(
... output_dir="rf_detr_finetuned_mobile_ui",
... num_train_epochs=5,
... bf16=True,
... per_device_train_batch_size=8,
... dataloader_num_workers=4,
... learning_rate=5e-5,
... lr_scheduler_type="cosine",
... weight_decay=1e-4,
... max_grad_norm=0.01,
... metric_for_best_model="eval_map",
... greater_is_better=True,
... load_best_model_at_end=True,
... eval_strategy="epoch",
... save_strategy="epoch",
... save_total_limit=2,
... remove_unused_columns=False,
... report_to="trackio",
... run_name="mobile-ui-detection",
... eval_do_concat_batches=False,
... push_to_hub=True,
... )最后,将所有内容整合起来,并调用 train():
>>> from transformers import Trainer
>>> trainer = Trainer(
... model=model,
... args=training_args,
... train_dataset=train_ds,
... eval_dataset=val_ds,
... processing_class=image_processor,
... data_collator=collate_fn,
... compute_metrics=eval_compute_metrics_fn,
... )
>>> trainer.train()| Epoch | 训练损失 | 验证损失 | 映射 | Map 50 | Map 75 | 映射小型 | 中等映射 | 映射大型 | 3 月 1 日 | 3 月 10 日 | 3月100日 | 3月 小 | 3月 中等 | Mar Large | 映射组 | 3月100组 | 图像映射 | Mar 100 图像 | 映射矩形 | Mar 100 矩形 | 文本映射 | 3月100 文本 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 无日志 | 9.9234 | 0.1303 | 0.2236 | 0.1478 | 0.0909 | 0.2030 | 0.2524 | 0.0421 | 0.2520 | 0.4683 | 0.3113 | 0.5607 | 0.6782 | 0.1244 | 0.5122 | 0.0958 | 0.5035 | 0.1285 | 0.4328 | 0.1725 | 0.4413 |
| 2 | 无日志 | 9.8472 | 0.1893 | 0.3017 | 0.2124 | 0.1347 | 0.2789 | 0.3038 | 0.0549 | 0.2961 | 0.5140 | 0.3433 | 0.5941 | 0.7406 | 0.1305 | 0.5423 | 0.1979 | 0.5578 | 0.1964 | 0.4648 | 0.2324 | 0.4437 |
| 3 | 无日志 | 9.6401 | 0.2275 | 0.3547 | 0.2657 | 0.1698 | 0.3336 | 0.3892 | 0.0611 | 0.3204 | 0.5270 | 0.3625 | 0.6143 | 0.7496 | 0.1602 | 0.5684 | 0.2617 | 0.5763 | 0.2249 | 0.4684 | 0.2631 | 0.4692 |
| 4 | 无日志 | 9.5770 | 0.2733 | 0.4068 | 0.3133 | 0.2100 | 0.3867 | 0.4343 | 0.0668 | 0.3456 | 0.5593 | 0.3875 | 0.6393 | 0.7725 | 0.2013 | 0.5941 | 0.3158 | 0.6065 | 0.2733 | 0.4998 | 0.3028 | 0.4756 |
| 5 | 10.3700 | 11.0500 | 0.2827 | 0.4193 | 0.2913 | 0.2021 | 0.2814 | 0.3763 | 0.0609 | 0.3403 | 0.5668 | 0.4138 | 0.5669 | 0.7317 | 0.2092 | 0.5979 | 0.3334 | 0.6295 | 0.2793 | 0.5245 | 0.3089 | 0.5151 |
如果在 training_args 中将 push_to_hub 设置为 True,训练检查点将被推送到 Hugging Face Hub。训练完成后,调用 push_to_hub() 方法将最终模型也推送到 Hub。
>>> trainer.push_to_hub()评估
>>> from pprint import pprint
>>> metrics = trainer.evaluate(eval_dataset=val_ds, metric_key_prefix="test")
>>> pprint(metrics)
{'test_loss': 11.05,
'test_map': 0.2827,
'test_map_50': 0.4193,
'test_map_75': 0.2913,
'test_map_group': 0.2092,
'test_map_image': 0.3334,
'test_map_large': 0.3763,
'test_map_medium': 0.2814,
'test_map_rectangle': 0.2793,
'test_map_small': 0.2021,
'test_map_text': 0.3089,
'test_mar_1': 0.0609,
'test_mar_10': 0.3403,
'test_mar_100': 0.5668,
'test_mar_100_group': 0.5979,
'test_mar_100_image': 0.6295,
'test_mar_100_rectangle': 0.5245,
'test_mar_100_text': 0.5151,
'test_mar_large': 0.7317,
'test_mar_medium': 0.5669,
'test_mar_small': 0.4138}通过增加训练轮数或调整 TrainingArguments 中的其他超参数,可以进一步提升这些结果。试试看吧!
推理
现在,您已经微调了模型、对其进行了评估,并将其上传到了 Hugging Face Hub,因此您可以将其用于推理。
>>> import torch
>>> from PIL import Image, ImageDraw
>>> from transformers import AutoImageProcessor, AutoModelForObjectDetection
>>> from datasets import load_dataset
>>> ds = load_dataset("merve/mobile-ui-design", split="train")
>>> image = ds[5]["image"].convert("RGB")从 Hugging Face Hub 加载模型和图像处理器(如果本次会话中已经训练过,可跳过):
>>> model_repo = "merve/rf_detr_finetuned_mobile_ui"
>>> image_processor = AutoImageProcessor.from_pretrained(model_repo)
>>> model = AutoModelForObjectDetection.from_pretrained(model_repo)
>>> model.eval()并检测边界框:
>>> with torch.no_grad():
... inputs = image_processor(images=[image], return_tensors="pt")
... outputs = model(**inputs)
... target_sizes = torch.tensor([[image.size[1], image.size[0]]])
... results = image_processor.post_process_object_detection(outputs, threshold=0.5, target_sizes=target_sizes)[0]
>>> for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
... box = [round(i, 2) for i in box.tolist()]
... print(
... f"Detected {model.config.id2label[label.item()]} with confidence "
... f"{round(score.item(), 3)} at location {box}"
... )
Detected text with confidence 0.727 at location [324.02, 340.55, 339.52, 359.12]
Detected rectangle with confidence 0.717 at location [39.97, 705.14, 335.93, 753.54]
Detected text with confidence 0.702 at location [199.94, 473.66, 213.41, 490.6]
Detected text with confidence 0.678 at location [153.14, 474.81, 165.33, 491.0]
Detected text with confidence 0.675 at location [262.67, 718.28, 281.44, 740.81]
Detected rectangle with confidence 0.655 at location [143.57, 242.51, 214.32, 274.26]
Detected text with confidence 0.653 at location [298.68, 637.77, 345.68, 656.26]让我们绘制结果:
>>> draw = ImageDraw.Draw(image)
>>> colors = {"group": "blue", "image": "green", "rectangle": "red", "text": "orange"}
>>> for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
... box = [round(i, 2) for i in box.tolist()]
... x, y, x2, y2 = tuple(box)
... label_name = model.config.id2label[label.item()]
... color = colors.get(label_name, "red")
... draw.rectangle((x, y, x2, y2), outline=color, width=2)
... draw.text((x, y), f"{label_name} {score:.2f}", fill=color)
>>> image
