---
license: mit
library_name: {{ library_name }}

pipeline_tag: {{ pipeline_tag }}

{% if base_model %}base_model: {{ base_model }}
{% endif %}tags:
{% for tag in tags %}  - {{ tag }}
{% endfor %}datasets:
{% for ds in datasets %}  - {{ ds }}
{% endfor %}
metrics:
  - oks
  - pck

{% if widget_examples %}widget:
{% for ex in widget_examples %}  - src: {{ ex.src }}
    example_title: {{ ex.example_title }}
{% endfor %}{% endif %}

model-index:
  - name: {{ repo_id }}
    results:
{% for task_result in metric_results %}      - task:
          type: {{ task_result.task_type }}
        dataset:
          type: {{ task_result.dataset_type }}
          name: {{ task_result.dataset_name }}
        metrics:
{% for m in task_result.metrics %}          - type: {{ m.type }}
            value: {{ m.value }}
{% endfor %}{% endfor %}---

# vehicle-keypoints

14-keypoint vehicle pose estimation on the CarFusion (CMU) dataset - four wheels, four head- and tail-lights, four roof corners, the exhaust, and a body-centre reference point per car (canonical CarFusion schema, Reddy et al., CVPR 2018). Main weights (`weights.pt`) are an Ultralytics YOLO26-pose checkpoint; a ViTPose-S top-down baseline is published under the `baseline/` subdirectory of this repo.

## Metrics (test set, n=12761)

{{ metrics_table }}

## Usage

```python
from huggingface_hub import hf_hub_download
from ultralytics import YOLO

ckpt = hf_hub_download(repo_id="{{ repo_id }}", filename="weights.pt")
model = YOLO(ckpt)
results = model.predict("car.jpg")
for r in results:
    for box, kpts, score in zip(r.boxes.xywh, r.keypoints.data, r.boxes.conf):
        print(box.tolist(), score.item(), len(kpts))
```

The baseline weights (ViTPose-S, HF `safetensors` format + processor config) live under the `baseline/` subdir of this repo and are loaded via `transformers` - see the [GitHub README]({{ github_url }}#baseline) for the inference snippet.

## Source

- **Code:** [{{ github_url }}]({{ github_url }})
- **Dataset:** [CarFusion](http://www.cs.cmu.edu/~ILIM/projects/IM/CarFusion/cvpr2018/index.html) - N. Dinesh Reddy, Minh Vo, Srinivasa Narasimhan, *"CarFusion: Combining Point Tracking and Part Detection for Dynamic 3D Reconstruction of Vehicles"*, CVPR 2018. © Carnegie Mellon University.
- **Keypoint order (14):** `right_front_wheel, left_front_wheel, right_back_wheel, left_back_wheel, right_front_headlight, left_front_headlight, right_back_headlight, left_back_headlight, exhaust, right_front_top, left_front_top, right_back_top, left_back_top, center` - naming follows the original CarFusion / Occlusion-Net reference (`dineshreddy91/Occlusion_Net/lib/data_loader/datasets/keypoint.py`).

## Intended use

Research and educational artifact demonstrating modern keypoint-detection pipelines on a non-human class. Not intended for any safety-critical, autonomous-driving, or surveillance deployment - the model is trained on a single academic dataset and has not been validated for production use.

## License

- **Code + weights:** MIT (see [`LICENSE`]({{ github_url }}/blob/main/LICENSE)).
- **Dataset:** CarFusion © Carnegie Mellon University - redistributed under the dataset's original terms; cite Reddy et al. 2018 if you use the weights for research.

> **Note:** This model card was generated from the [ml-project-template](https://github.com/kiselyovd/ml-project-template) scaffold.