Object detection is the computer vision task of detecting instances (such as humans, buildings, or cars) in an image. Object detection models receive an image as input and output coordinates of the bounding boxes and associated labels of the detected objects. An image can contain multiple objects, each with its own bounding box and a label (e.g. it can have a car and a building), and each object can be present in different parts of an image (e.g. the image can have several cars). This task is commonly used in autonomous driving for detecting things like pedestrians, road signs, and traffic lights. Other applications include counting objects in images, image search, and more.
In this guide, you will learn how to:
Finetune DETR, a model that combines a convolutional backbone with an encoder-decoder Transformer, on the CPPE-5 dataset.
Use your finetuned model for inference.
The task illustrated in this tutorial is supported by the following model architectures:
You’ll use 🌍 Datasets to load a dataset from the BOINC AI Hub, 🌍 Transformers to train your model, and albumentations to augment the data. timm is currently required to load a convolutional backbone for the DETR model.
We encourage you to share your model with the community. Log in to your BOINC AI account to upload it to the Hub. When prompted, enter your token to log in:
Copied
>>> from boincai_hub import notebook_login
>>> notebook_login()
Load the CPPE-5 dataset
The CPPE-5 dataset contains images with annotations identifying medical personal protective equipment (PPE) in the context of the COVID-19 pandemic.
The examples in the dataset have the following fields:
image_id: the example image id
image: a PIL.Image.Image object containing the image
width: width of the image
height: height of the image
objects: a dictionary containing bounding box metadata for the objects in the image:
id: the annotation id
area: the area of the bounding box
bbox: the object’s bounding box (in the COCO format )
category: the object’s category, with possible values including Coverall (0), Face_Shield (1), Gloves (2), Goggles (3) and Mask (4)
You may notice that the bbox field follows the COCO format, which is the format that the DETR model expects. However, the grouping of the fields inside objects differs from the annotation format DETR requires. You will need to apply some preprocessing transformations before using this data for training.
To get an even better understanding of the data, visualize an example in the dataset.
Copied
>>> import numpy as np
>>> import os
>>> from PIL import Image, ImageDraw
>>> image = cppe5["train"][0]["image"]
>>> annotations = cppe5["train"][0]["objects"]
>>> draw = ImageDraw.Draw(image)
>>> categories = cppe5["train"].features["objects"].feature["category"].names
>>> id2label = {index: x for index, x in enumerate(categories, start=0)}
>>> label2id = {v: k for k, v in id2label.items()}
>>> for i in range(len(annotations["id"])):
... box = annotations["bbox"][i]
... class_idx = annotations["category"][i]
... x, y, w, h = tuple(box)
... draw.rectangle((x, y, x + w, y + h), outline="red", width=1)
... draw.text((x, y), id2label[class_idx], fill="white")
>>> image
To visualize the bounding boxes with associated labels, you can get the labels from the dataset’s metadata, specifically the category field. You’ll also want to create dictionaries that map a label id to a label class (id2label) and the other way around (label2id). You can use them later when setting up the model. Including these maps will make your model reusable by others if you share it on the BOINC AI Hub.
As a final step of getting familiar with the data, explore it for potential issues. One common problem with datasets for object detection is bounding boxes that “stretch” beyond the edge of the image. Such “runaway” bounding boxes can raise errors during training and should be addressed at this stage. There are a few examples with this issue in this dataset. To keep things simple in this guide, we remove these images from the data.
Copied
>>> remove_idx = [590, 821, 822, 875, 876, 878, 879]
>>> keep = [i for i in range(len(cppe5["train"])) if i not in remove_idx]
>>> cppe5["train"] = cppe5["train"].select(keep)
Preprocess the data
To finetune a model, you must preprocess the data you plan to use to match precisely the approach used for the pre-trained model. AutoImageProcessor takes care of processing image data to create pixel_values, pixel_mask, and labels that a DETR model can train with. The image processor has some attributes that you won’t have to worry about:
image_mean = [0.485, 0.456, 0.406 ]
image_std = [0.229, 0.224, 0.225]
These are the mean and standard deviation used to normalize images during the model pre-training. These values are crucial to replicate when doing inference or finetuning a pre-trained image model.
Instantiate the image processor from the same checkpoint as the model you want to finetune.
Before passing the images to the image_processor, apply two preprocessing transformations to the dataset:
Augmenting images
Reformatting annotations to meet DETR expectations
First, to make sure the model does not overfit on the training data, you can apply image augmentation with any data augmentation library. Here we use Albumentations … This library ensures that transformations affect the image and update the bounding boxes accordingly. The 🌍 Datasets library documentation has a detailed guide on how to augment images for object detection, and it uses the exact same dataset as an example. Apply the same approach here, resize each image to (480, 480), flip it horizontally, and brighten it:
The image_processor expects the annotations to be in the following format: {'image_id': int, 'annotations': List[Dict]}, where each dictionary is a COCO object annotation. Let’s add a function to reformat annotations for a single example:
Apply this preprocessing function to the entire dataset using 🌍Datasets with_transform method. This method applies transformations on the fly when you load an element of the dataset.
At this point, you can check what an example from the dataset looks like after the transformations. You should see a tensor with pixel_values, a tensor with pixel_mask, and labels.
You have successfully augmented the individual images and prepared their annotations. However, preprocessing isn’t complete yet. In the final step, create a custom collate_fn to batch images together. Pad images (which are now pixel_values) to the largest image in a batch, and create a corresponding pixel_mask to indicate which pixels are real (1) and which are padding (0).
You have done most of the heavy lifting in the previous sections, so now you are ready to train your model! The images in this dataset are still quite large, even after resizing. This means that finetuning this model will require at least one GPU.
When loading the model from the same checkpoint that you used for the preprocessing, remember to pass the label2id and id2label maps that you created earlier from the dataset’s metadata. Additionally, we specify ignore_mismatched_sizes=True to replace the existing classification head with a new one.
Copied
>>> from transformers import AutoModelForObjectDetection
>>> model = AutoModelForObjectDetection.from_pretrained(
... checkpoint,
... id2label=id2label,
... label2id=label2id,
... ignore_mismatched_sizes=True,
... )
In the TrainingArguments use output_dir to specify where to save your model, then configure hyperparameters as you see fit. It is important you do not remove unused columns because this will drop the image column. Without the image column, you can’t create pixel_values. For this reason, set remove_unused_columns to False. If you wish to share your model by pushing to the Hub, set push_to_hub to True (you must be signed in to BOINC AI to upload your model).
If you have set push_to_hub to True in the training_args, the training checkpoints are pushed to the BOINC AI Hub. Upon training completion, push the final model to the Hub as well by calling the push_to_hub() method.
Copied
>>> trainer.push_to_hub()
Evaluate
Object detection models are commonly evaluated with a set of COCO-style metrics. You can use one of the existing metrics implementations, but here you’ll use the one from torchvision to evaluate the final model that you pushed to the Hub.
To use the torchvision evaluator, you’ll need to prepare a ground truth COCO dataset. The API to build a COCO dataset requires the data to be stored in a certain format, so you’ll need to save images and annotations to disk first. Just like when you prepared your data for training, the annotations from the cppe5["test"] need to be formatted. However, images should stay as they are.
The evaluation step requires a bit of work, but it can be split in three major steps. First, prepare the cppe5["test"] set: format the annotations and save the data to disk.
Copied
>>> import json
>>> # format annotations the same as for training, no need for data augmentation
>>> def val_formatted_anns(image_id, objects):
... annotations = []
... for i in range(0, len(objects["id"])):
... new_ann = {
... "id": objects["id"][i],
... "category_id": objects["category"][i],
... "iscrowd": 0,
... "image_id": image_id,
... "area": objects["area"][i],
... "bbox": objects["bbox"][i],
... }
... annotations.append(new_ann)
... return annotations
>>> # Save images and annotations into the files torchvision.datasets.CocoDetection expects
>>> def save_cppe5_annotation_file_images(cppe5):
... output_json = {}
... path_output_cppe5 = f"{os.getcwd()}/cppe5/"
... if not os.path.exists(path_output_cppe5):
... os.makedirs(path_output_cppe5)
... path_anno = os.path.join(path_output_cppe5, "cppe5_ann.json")
... categories_json = [{"supercategory": "none", "id": id, "name": id2label[id]} for id in id2label]
... output_json["images"] = []
... output_json["annotations"] = []
... for example in cppe5:
... ann = val_formatted_anns(example["image_id"], example["objects"])
... output_json["images"].append(
... {
... "id": example["image_id"],
... "width": example["image"].width,
... "height": example["image"].height,
... "file_name": f"{example['image_id']}.png",
... }
... )
... output_json["annotations"].extend(ann)
... output_json["categories"] = categories_json
... with open(path_anno, "w") as file:
... json.dump(output_json, file, ensure_ascii=False, indent=4)
... for im, img_id in zip(cppe5["image"], cppe5["image_id"]):
... path_img = os.path.join(path_output_cppe5, f"{img_id}.png")
... im.save(path_img)
... return path_output_cppe5, path_anno
Next, prepare an instance of a CocoDetection class that can be used with cocoevaluator.
>>> import evaluate
>>> from tqdm import tqdm
>>> model = AutoModelForObjectDetection.from_pretrained("devonho/detr-resnet-50_finetuned_cppe5")
>>> module = evaluate.load("ybelkada/cocoevaluate", coco=test_ds_coco_format.coco)
>>> val_dataloader = torch.utils.data.DataLoader(
... test_ds_coco_format, batch_size=8, shuffle=False, num_workers=4, collate_fn=collate_fn
... )
>>> with torch.no_grad():
... for idx, batch in enumerate(tqdm(val_dataloader)):
... pixel_values = batch["pixel_values"]
... pixel_mask = batch["pixel_mask"]
... labels = [
... {k: v for k, v in t.items()} for t in batch["labels"]
... ] # these are in DETR format, resized + normalized
... # forward pass
... outputs = model(pixel_values=pixel_values, pixel_mask=pixel_mask)
... orig_target_sizes = torch.stack([target["orig_size"] for target in labels], dim=0)
... results = im_processor.post_process(outputs, orig_target_sizes) # convert outputs of model to COCO api
... module.add(prediction=results, reference=labels)
... del batch
>>> results = module.compute()
>>> print(results)
Accumulating evaluation results...
DONE (t=0.08s).
IoU metric: bbox
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.352
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.681
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.292
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.168
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.208
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.429
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.274
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.484
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.501
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.191
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.323
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.590
These results can be further improved by adjusting the hyperparameters in TrainingArguments. Give it a go!
Inference
Now that you have finetuned a DETR model, evaluated it, and uploaded it to the BOINC AI Hub, you can use it for inference. The simplest way to try out your finetuned model for inference is to use it in a Pipeline. Instantiate a pipeline for object detection with your model, and pass an image to it:
