---
license: mit
tags:
  - pose-estimation
  - 6d-pose
  - vision-transformer
  - spacecraft
  - space
  - vit
datasets:
  - SPEED
pipeline_tag: image-classification
---

# FastPose-ViT: Pretrained Weights

Pretrained weights for **[FastPose-ViT](https://github.com/PierreAncey/FastPose-ViT)**, a Vision Transformer pipeline for real-time 6D spacecraft pose estimation.

## Available Weights

All models are trained on the **SPEED** dataset.

| File | Model | Task | Input Resolution |
|------|-------|------|-----------------|
| `vit_b_16_384.pth` | ViT-B/16-384 | Pose estimation (6D) | 384x384 |
| `vit_b_16.pth` | ViT-B/16 | Pose estimation (6D) | 224x224 |
| `small.pth` | LW-DETR Small | Object detection (bbox) | 512x512 |

## Usage

1. Clone the repository:
```bash
git clone https://github.com/PierreAncey/FastPose-ViT.git
cd FastPose-ViT
```

2. Download weights and place them in a `weights/` directory.

3. Run evaluation on the SPEED dataset:
```bash
DATASET=SPEED_FIXED && \
python3 src/evaluate.py \
  --model_weights weights/vit_b_16_384.pth \
  --rotation_format matrix \
  --num_hidden_layers 0 \
  --hidden_layer_dim 0 \
  --nb_class_tokens 1 \
  --batch_size 8 \
  --vit_model vit_b_16_384 \
  --dataset SPEED \
  --dataset_root_dir $DATASET \
  --num_workers 8 \
  --merge_outputs \
  --no_mlp
```

4. Run the object detector:
```bash
DATASET=SPEED_FIXED && \
python3 object_detector/evaluate.py \
  --dataset_root_dir $DATASET \
  --model_variant small \
  --model_weights weights/small.pth
```

## Model Details

- **Pose estimator**: ViT backbone with direct 6D pose regression (rotation matrix + translation vector). Uses 6D continuous rotation representation with Gram-Schmidt orthogonalization.
- **Object detector**: LW-DETR (Lightweight DETR) fine-tuned from COCO-pretrained weights for single-class spacecraft detection. Provides bounding boxes as preprocessing for the pose estimator.

## Citation

```bibtex
@InProceedings{Ancey_2026_WACV,
    author    = {Ancey, Pierre and Price, Andrew and Javed, Saqib and Salzmann, Mathieu},
    title     = {FastPose-ViT: A Vision Transformer for Real-Time Spacecraft Pose Estimation},
    booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
    month     = {March},
    year      = {2026},
    pages     = {7873-7882}
}
```

## License

MIT License. See the [repository](https://github.com/PierreAncey/FastPose-ViT) for details.