Detr Resnet 50

🚀 DETR (End-to-End Object Detection) model with ResNet-50 backbone

The DETR model is trained end-to-end on COCO 2017 for object detection, offering an efficient solution for identifying objects in images.

🚀 Quick Start

The DETR model is an end-to-end object detection model trained on the COCO 2017 dataset. It simplifies the object detection process by leveraging a transformer architecture. You can use it for various object detection tasks right away.

✨ Features

• End-to-End Training: Trained directly on the COCO 2017 dataset, eliminating the need for complex post-processing steps.
• Transformer Architecture: Utilizes an encoder-decoder transformer with a convolutional backbone for effective object detection.
• Object Queries: Employs object queries to detect objects in an image, with 100 queries set for the COCO dataset.

📦 Installation

The code examples provided assume you have the necessary libraries installed. You can install them using the following command:

pip install transformers torch pillow requests

💻 Usage Examples

Basic Usage

from transformers import DetrImageProcessor, DetrForObjectDetection
import torch
from PIL import Image
import requests

url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)

# you can specify the revision tag if you don't want the timm dependency
processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50", revision="no_timm")
model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50", revision="no_timm")

inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)

# convert outputs (bounding boxes and class logits) to COCO API
# let's only keep detections with score > 0.9
target_sizes = torch.tensor([image.size[::-1]])
results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[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}"
    )

This should output:

Detected remote with confidence 0.998 at location [40.16, 70.81, 175.55, 117.98]
Detected remote with confidence 0.996 at location [333.24, 72.55, 368.33, 187.66]
Detected couch with confidence 0.995 at location [-0.02, 1.15, 639.73, 473.76]
Detected cat with confidence 0.999 at location [13.24, 52.05, 314.02, 470.93]
Detected cat with confidence 0.999 at location [345.4, 23.85, 640.37, 368.72]

📚 Documentation

Model description

The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.

The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.

Intended uses & limitations

You can use the raw model for object detection. See the model hub to look for all available DETR models.

Training data

The DETR model was trained on COCO 2017 object detection, a dataset consisting of 118k/5k annotated images for training/validation respectively.

Training procedure

Preprocessing

The exact details of preprocessing of images during training/validation can be found here.

Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).

Training

The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).

Evaluation results

This model achieves an AP (average precision) of 42.0 on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper.

🔧 Technical Details

• Model Type: Encoder-decoder transformer with a convolutional backbone.
• Loss Function: Bipartite matching loss, using cross-entropy for classes and a combination of L1 and generalized IoU loss for bounding boxes.
• Optimization: The Hungarian matching algorithm is used for optimal one-to-one mapping between queries and annotations.

📄 License

This model is released under the Apache-2.0 license.

BibTeX entry and citation info

@article{DBLP:journals/corr/abs-2005-12872,
  author    = {Nicolas Carion and
               Francisco Massa and
               Gabriel Synnaeve and
               Nicolas Usunier and
               Alexander Kirillov and
               Sergey Zagoruyko},
  title     = {End-to-End Object Detection with Transformers},
  journal   = {CoRR},
  volume    = {abs/2005.12872},
  year      = {2020},
  url       = {https://arxiv.org/abs/2005.12872},
  archivePrefix = {arXiv},
  eprint    = {2005.12872},
  timestamp = {Thu, 28 May 2020 17:38:09 +0200},
  biburl    = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}