---
license: apache-2.0
tags:
- model_hub_mixin
- pytorch_model_hub_mixin
language:
- en
pipeline_tag: keypoint-detection
---

# DETRPose-L-COCO

DETRPose-L-COCO is a real-time object detection model introduced in the paper [DETRPose: Real-Time End-to-End Multi-Person Pose Estimation via Modified Transformer Decoder and Novel Denoising Keypoints](https://huggingface.co/papers/2506.13027). 

- **Repository:** [https://github.com/SebastianJanampa/DETRPose](https://github.com/SebastianJanampa/DETRPose)
- **Paper:** [https://huggingface.co/papers/2506.13027](https://huggingface.co/papers/2506.13027)

## 📝 Model Description
DETRPose introduces the first real-time end-to-end framework for multi-person pose estimation. 
By leveraging the hybrid encoder from RT-DETR and the lightweight decoder architecture of D-FINE, DETRPose achieves low-latency inference without sacrificing accuracy.
The model introduces two primary methodological advancements:

* **Pose-LQE Layer:** A specialized head designed to improve confidence scores.
* **Advanced Training Paradigm:** Incorporates Denoising Keypoints and a custom Keypoint Similarity Varifocal loss function, ensuring robust learning and superior localization performance.


| Model | Dataset | AP | #Params | Latency | GFLOPs |
| :---: | :---: | :---: | :---: | :---: | :---: |
| **DETRPose-L** | COCO | 72.5 | 32.8 M | 9.50 ms | 107.1 |


## 🚀 Installation

To use this model, you need to install the inference-ready branch of the DETRPose repository.
You can directly install the inference-ready branch using `pip`:
```shell
pip install git+https://github.com/SebastianJanampa/DETRPose.git@inference_only
````

## 💻 Usage
This branch is designed to be easy to use for inference. 
Here is a quick example of how to load a model and run it on a live webcam feed.
```python
from detrpose import DETR

# Initialization
model = DETR(model='detrpose_hgnetv2_l')

# Inference
model(source=0) # inference on a webcam
```


## 📜 Citation
If you use `DETRPose` or its methods in your work, please cite the following BibTeX entries:

```latex
@misc{janampa2025detrpose,
      title={DETRPose: Real-time end-to-end transformer model for multi-person pose estimation}, 
      author={Sebastian Janampa and Marios Pattichis},
      year={2025},
      eprint={2506.13027},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2506.13027}, 
}
```

## 🙏 Acknowledgement
This work was supported in part by [Lambda.ai](https://lambda.ai).

Our work is built upon [DEIM](https://github.com/Intellindust-AI-Lab/DEIM/tree/main), [D-FINE](https://github.com/Peterande/D-FINE), [Detectron2](https://github.com/facebookresearch/detectron2/tree/main), and [GroupPose](https://github.com/Michel-liu/GroupPose/tree/main).

✨ Feel free to reach out if you have any questions! ✨

<div align="left">
  <a href="https://lambda.ai" target="_blank">
  	<img src="./assets/lambda_logo2.png" width=500 >
  </a>
</div>