# PG-StreetParse-M

PG-StreetParse-M is a public instance segmentation model by Pictograph Research on Pictograph, trained on the A2D2 dataset and exportable to ONNX for inference. Reported mAP of 0.189.

## Overview

- Type: instance segmentation model
- Task: Segmentation (https://pictograph.io/explore/task/segmentation)
- Architecture: RF-DETR Seg Medium
- Owner: Pictograph Research
- Trained on: A2D2 (https://pictograph.io/datasets/pictograph-research/a2d2)
- License: Creative Commons Attribution-NoDerivatives 4.0 (https://creativecommons.org/licenses/by-nd/4.0/)
- Export: ONNX
- Stars: 0
- Page: https://pictograph.io/models/pictograph-research/pg-streetparse-m
- Hugging Face: https://huggingface.co/pictograph/pg-streetparse-m

## Evaluation metrics

- mAP: 18.9%
- mAP50: 36.8%

## About

# PG-StreetParse-M

> An instance segmentation model (RF-DETR Seg Medium), trained on 6 classes, scoring 18.9% mAP.

**[Pictograph Research](/org/pictograph-research)** · Instance Segmentation · RF-DETR Seg Medium · FP32 · [CC-BY-ND-4.0](https://creativecommons.org/licenses/by-nd/4.0/) · Public

## Overview

This is an instance segmentation model built on **RF-DETR Seg Medium**, trained on the [A2D2](/datasets/pictograph-research/a2d2) dataset. It expects **432x432** input and runs in FP32 precision. It predicts car, pedestrian, truck, small vehicle, utility vehicle, bicycle.

## Performance

Headline: **18.9% mAP**.

| Metric | Value |
| :--- | ---: |
| mAP | 18.9% |
| mAP@50 | 36.8% |

## Classes

Class index matches the model's output order.

| # | Class |
| ---: | :--- |
| 0 | car |
| 1 | pedestrian |
| 2 | truck |
| 3 | small vehicle |
| 4 | utility vehicle |
| 5 | bicycle |

## Training

| Setting | Value |
| :--- | :--- |
| Base dataset | [A2D2](/datasets/pictograph-research/a2d2) |
| Export | A2D2-base-sample |
| Dataset size | 12,758 images |
| Input size | 432x432 |
| Epochs | 10 |
| Batch size | 4 |
| Learning rate | 0.0002 |
| Model size | medium |
| Precision | FP32 |
| Version | 1.0.0 |

## Use this model

Run inference with the Pictograph SDK:

```python
from pictograph import Client, get_model, InstanceSegmentationModel, InstanceSegmentationResult

client = Client(api_key="pk_live_...")  # Settings -> API Keys

# Downloads + caches the weights on first use
model: InstanceSegmentationModel = get_model(
    name="PG-StreetParse-M",
    task="instance_segmentation",
    format="onnx",
    precision="fp32",
    device="cpu",
    client=client,
)
result: InstanceSegmentationResult = model.predict(image="photo.jpg")
print(result)
```

```python
# Fully offline (no API key) once you have the weights + config.json
from pictograph import load_model, InstanceSegmentationModel, InstanceSegmentationResult

model: InstanceSegmentationModel = load_model(
    "/path/to/weights.onnx",
    "/path/to/config.json",
    task="instance_segmentation",
    format="onnx",
)
result: InstanceSegmentationResult = model.predict(image="photo.jpg")
```

- Full inference guide: [local inference docs](https://pictograph.io/docs/local-inference).
- REST + CLI: [models API reference](https://pictograph.io/docs/api-reference/models).
- Deploy it as an always-on endpoint: [deployments](https://pictograph.io/docs/deployments).

## License

Released under [CC-BY-ND-4.0](https://creativecommons.org/licenses/by-nd/4.0/), the license of its training data, [A2D2](https://a2d2-dataset.github.io). Commercial use is permitted with attribution: when you use this model, please credit **Audi AG (Geyer et al., 2020)**.

## Citation

This model uses the RF-DETR Seg Medium architecture:

```bibtex
@inproceedings{robinson2026rfdetr,
  title     = {RF-DETR: Real-Time Detection Transformer},
  author    = {Robinson, Isaac and Robicheaux, Peter and Popov, Matvei and Ramanan, Deva and Peri, Neehar},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year      = {2026},
  url       = {https://arxiv.org/abs/2511.09554}
}
```

It was trained on A2D2, from:

```bibtex
@article{geyer2020a2d2,
  title = {{A2D2}: {Audi Autonomous Driving Dataset}},
  author = {Geyer, Jakob and Kassahun, Yohannes and Mahmudi, Mentar and Ricou, Xavier and Durgesh, Rupesh and Chung, Andrew S. and Hauswald, Lorenz and Pham, Viet Hoang and M{\"u}hlegg, Maximilian and Dorn, Sebastian and Fernandez, Tiffany and J{\"a}nicke, Martin and Mirashi, Sudesh and Savani, Chiragkumar and Sturm, Martin and Vorobiov, Oleksandr and Oelker, Martin and Garreis, Sebastian and Schuberth, Peter},
  journal = {arXiv preprint arXiv:2004.06320},
  year = {2020}
}
```


---

Free to view, and to run in your browser on your own GPU (an image, a video or your camera; no sign-in): https://app.pictograph.io/explore/models/pictograph-research/pg-streetparse-m?run=1

Sign in (free) to fork this model, run batch inference, or deploy it as a live endpoint; this link opens it in the app afterwards: https://app.pictograph.io/signup?next=%2Fexplore%2Fmodels%2Fpictograph-research%2Fpg-streetparse-m
