---
task_categories:
- text-to-3d
- image-to-3d
- image-text-to-image
- any-to-any
tags:
- Camera
- 3D Vision
- Spatial AI
- Physical AI
- World Model
- Camera Parameter
- AP-10K
- Generation

---

# AP-10K-Camera

![camera map collage](analysis/AP-10K-Camera.png)

Per-image camera parameter annotations for the **AP-10K** dataset
(an animal pose estimation benchmark; 10,015 images across 54 species from 23 families), captioned by the [**Puffin-World**](https://github.com/KangLiao929/Puffin) model. More captioned datasets are provided in our [**Puffin-16M**](https://kangliao929.github.io/projects/puffin-16m/) website.

The collage above visualizes the camera maps on sample images — each
pair shows the **up field** (green arrows: the projected gravity-up direction)
and the **latitude field** (colored contours: angle above/below the horizon).

## Format

One `.tar` per shard (`ap10k_0000.tar` … ), each containing one `.json` per
image whose name matches the source image stem.

Each JSON holds the predicted monocular camera parameters:

| Field | Meaning | Unit |
|-------|---------|------|
| `roll`  | camera roll        | radians |
| `pitch` | camera pitch       | radians |
| `vfov`  | vertical field-of-view | radians |
| `k1`    | radial distortion coefficient | – |
| `parse_ok` | whether the model output parsed within valid ranges | bool |

Example:

```json
{"roll": 0.0123, "pitch": -0.1604, "vfov": 0.4620, "k1": 0.0000, "parse_ok": true}
```

## Camera Parameter Distributions

Histograms of the predicted roll / pitch / vertical-FoV over the whole dataset
(proportion of valid samples per 10° bin; `parse_ok=False` excluded).

![ap10k camera stats](analysis/ap10k_camera_stats.png)

| split | roll μ / med / σ | pitch μ / med / σ | FoV μ / med / σ |
|-------|------------------|-------------------|-----------------|
| all (9,999) | 0.4° / 0.2° / 8.1° | −9.2° / −6.1° / 13.9° | 26.5° / 23.2° / 8.3° |

- **Roll** is centered at 0° with a moderate spread (σ ≈ 8°).
- **Pitch** is markedly negative (median ≈ −6°, long downward tail) — animals are
  often photographed from a slightly elevated, downward-looking viewpoint.
- **FoV** is fairly narrow (median ≈ 23°) — telephoto lenses for wildlife shots.

If you'd like a dataset with a more diverse and uniform distribution of camera parameters, please refer to our [Puffin-4M](https://huggingface.co/datasets/KangLiao/Puffin-4M) and [Puffin-16M](https://huggingface.co/datasets/KangLiao/Puffin-16M) datasets.

### Dataset Download
You can download the entire dataset using the following command:
```bash
hf download KangLiao/AP-10K-Camera --repo-type dataset
```

### Caption Pipeline
Beyond this captioned dataset, we also release **a complete captioning pipeline** for annotating camera parameters for arbitrary datasets, analyzing camera parameter distributions, and visualizing the corresponding camera maps. The pipeline is available in our [GitHub repository](https://github.com/KangLiao929/Puffin).


### Citation
If you find the captioned dataset useful for your research or applications, please cite our paper using the following BibTeX:

```bibtex
  @article{liao2025puffin,
    title={Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation},
    author={Liao, Kang and Wu, Size and Wu, Zhonghua and Jin, Linyi and Wang, Chao and Wang, Yikai and Wang, Fei and Li, Wei and Loy, Chen Change},
    journal={arXiv preprint arXiv:2510.08673},
    year={2025}
  }
```