# RiO-DETRv2
RiO-DETRv2 is a real-time oriented-box DETR. It predicts four box sides as distributions in the box's own frame, refines the angle coarse to fine and scores each box with an angle-aware quality. LibreYOLO ports it inference-only, in five sizes, with DOTA-v1.0 and DIOR-R checkpoints.
Tasks: Oriented boxes. Install: pip install libreyolo.
Verified against LibreYOLO v1.8.0.

## Install

RiO-DETRv2 needs no optional extra.

```bash
pip install libreyolo
```

## Predict

Weights download from the LibreYOLO Hugging Face mirrors on first use. The
models are trained on aerial imagery, so use an overhead photo.

**Python**

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreRiODETRv2s-obb.pt")        # DOTA-v1.0, 15 classes
result = model("aerial.jpg", save=True)

print(result.obb.xywhr)   # (N, 5) centre, size and angle
print(result.obb.conf, result.obb.cls)
```

**DIOR-R**

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreRiODETRv2s-obb-diorr.pt")  # DIOR-R, 20 classes
result = model("aerial.jpg")
```

**CLI**

```bash
libreyolo predict model=LibreRiODETRv2s-obb.pt source=aerial.jpg save=True
```

`result.obb` holds the rotated boxes: `xywhr` (centre, width, height and
angle), `conf` and `cls`, with class names from the checkpoint. See the
[oriented detection](/docs/tasks/oriented-detection) task page for the
payload and plotting.

## Variants

Five sizes, `n`, `s`, `m`, `l` and `x`, each with two checkpoints:

| Checkpoint | Dataset | Classes | Canvas |
| --- | --- | ---: | ---: |
| `LibreRiODETRv2<size>-obb.pt` | DOTA-v1.0 | 15 | 1024 px |
| `LibreRiODETRv2<size>-obb-diorr.pt` | DIOR-R | 20 | 800 px |

The native canvas is read from the checkpoint. Raw outputs and decoded
rotated boxes match the upstream Apache-2.0 release exactly (max absolute
difference 0) for all ten checkpoints on CPU.

RiO-DETRv2 is inference-only. `train()` raises: the upstream recipe (DINOv3
feature distillation, OBB Hungarian matching, boundary and angle
distribution losses) is not ported. For trainable oriented-box detection use
[GTR](/docs/models/gtr) or [YOLO-NAS](/docs/models/yolo-nas).

## Export

| Task | ONNX | TorchScript | ExecuTorch | TensorRT | OpenVINO | Paddle | MNN | RKNN | Axelera | ncnn | TFLite | CoreML | Core AI |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Oriented boxes | yes | yes | yes | yes | yes |  |  |  |  |  |  |  |  |

A "yes" means the export is supported. An empty cell means the exporter refuses that combination.

ONNX and TorchScript exports round-trip: the exported file loads back through
`LibreYOLO()` and returns the same `Results`.

**Python**

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreRiODETRv2s-obb.pt")
model.export(format="onnx")
model.export(format="torchscript")
```

**CLI**

```bash
libreyolo export model=LibreRiODETRv2s-obb.pt format=onnx
```

## Checkpoints

| File | Input (px) | Task | Weights license |
| --- | --- | --- | --- |
| `LibreRiODETRv2l-obb.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2l-obb-diorr.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2m-obb.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2m-obb-diorr.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2n-obb.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2n-obb-diorr.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2s-obb.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2s-obb-diorr.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2x-obb.pt` |  | Oriented boxes | apache-2.0 |
| `LibreRiODETRv2x-obb-diorr.pt` |  | Oriented boxes | apache-2.0 |

## Licensing

Check the license on the Hugging Face repository of the specific weights you download. That repository is authoritative and licenses are not always uniform across a family. This is a description of the licenses involved, not legal advice.

- Original work: RiO-DETRv2, RiO-DETRv2 authors (RicePasteM)
- Upstream license: Apache-2.0
- Upstream source: https://github.com/RicePasteM/RiO-DETRv2
- LibreYOLO code: Apache-2.0
- Weights: Apache-2.0, republished at https://huggingface.co/LibreYOLO
- Interpretation: Apache-2.0 is a permissive license, so the RiO-DETRv2 code and checkpoints can be used in commercial and closed-source products. Redistributing them means keeping the license text and notices, and the license grants a patent license. About 33 minutes after its initial commit (ab8b22df5d800a927774c8c67e77fd29c0fe6967), upstream changed its LICENSE file to AGPL-3.0; later commits change only the README and LICENSE, not the code. LibreYOLO's port relies on the Apache-2.0 grant of that initial release, uses nothing from later commits, and preserves the Apache-2.0 snapshot at LibreYOLO/RiO-DETRv2 (tag apache-2.0-ab8b22d). The adapted decoder keeps Apache-2.0 with its RT-DETR, D-FINE, DEIM and RT-DETRv4 copyright lines (libreyolo/models/riodetrv2/LICENSE and NOTICE); the HGNetv2 backbone, encoder and decoder blocks reuse LibreYOLO's dfine and rtdetrv2 families, and the wiring around them is LibreYOLO's MIT code. The official checkpoints (Hugging Face RicePasteM/RiO-DETRv2 at revision 5f578f57929aded1962ebf8c5584a4771461124f, tagged apache-2.0) are mirrored as LibreYOLO/LibreRiODETRv2{n,s,m,l,x}-obb (DOTA-v1.0) and -obb-diorr (DIOR-R) with the upstream LICENSE and NOTICE; conversion renames keys and drops a training-only projector, learned tensors are unchanged. The DOTA-v1.0 and DIOR-R training sets carry their own terms.

RiO-DETRv2 is by RicePasteM
([RicePasteM/RiO-DETRv2](https://github.com/RicePasteM/RiO-DETRv2)). Code and
weights are Apache-2.0, so commercial use is fine. Upstream changed its
LICENSE file to AGPL-3.0 about 33 minutes after its initial commit, with no
code changes. LibreYOLO's port uses only that initial Apache-2.0 release
(commit `ab8b22d`) and preserves the snapshot at `LibreYOLO/RiO-DETRv2`. The
official checkpoints are mirrored as `LibreYOLO/LibreRiODETRv2{n,s,m,l,x}-obb`
and `-obb-diorr`, with keys renamed and learned tensors unchanged. The
DOTA-v1.0 and DIOR-R training sets carry their own terms.

## Citation

```bibtex
@misc{hu2026riodetrv2geometryawareuncertaintymodeling,
      title={RiO-DETRv2: Geometry-Aware Uncertainty Modeling Meets Real-Time Oriented Object Detection Transformer}, 
      author={Zhangchi Hu and Zilong He and Yifan Zhao and Xiangyu Yu and Jie Chen and Peixi Wu and Yansong Peng and Hebei Li and Xiaoyan Sun},
      year={2026},
      eprint={2610.29159},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2610.29159}, 
}
```

Copied from https://arxiv.org/bibtex/2610.29159
