Metadata-Version: 2.4
Name: EquiForge
Version: 0.3.0
Summary: A toolkit for equirectangular image processing and conversions.
Author-email: Mikkel Kappel Persson <mikkelkp@hotmail.com>
Project-URL: Homepage, https://github.com/mikkelkappelpersson/equiforge
Keywords: equirectangular,panorama,image processing,360
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: License :: OSI Approved :: Apache Software License
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.20.0
Requires-Dist: torch>=2.4.0
Requires-Dist: Pillow>=8.0.0
Provides-Extra: dev
Requires-Dist: pytest>=6.0.0; extra == "dev"
Requires-Dist: jupyter>=1.0.0; extra == "dev"
Requires-Dist: matplotlib>=3.4.0; extra == "dev"
Requires-Dist: matplotlib>=3.4.0; extra == "dev"
Dynamic: license-file

<p align="center">
    <img src="src/Logo+Name.svg" alt="EquiForge Logo" width="500"/>
</p>

<h4 align="center">
    A precision toolkit for equirectangular image processing and conversions
</h4>

<div align="center">
<a href="https://badge.fury.io/py/equiforge"><img src="https://badge.fury.io/py/equiforge.svg" alt="PyPI version"></a>
<a href="https://pypi.org/project/equiforge"><img src="https://img.shields.io/pypi/pyversions/equiforge"></a>
  <a href="https://github.com/MikkelKappelPersson/EquiForge/actions"><img src="https://github.com/MikkelKappelPersson/EquiForge/actions/workflows/python-package-tests.yml/badge.svg"></a>
  <a href="https://github.com/MikkelKappelPersson/EquiForge/blob/main/LICENSE"><img alt="GitHub license" src="https://img.shields.io/github/license/mikkelkappelpersson/equiforge"></a>
</div>

## Features

- Convert perspective images to equirectangular projection (`pers2equi`)
- Convert equirectangular images to perspective views (`equi2pers`)
- Runs on CPU and CUDA via torch
- HDR support with `output_dtype="float32"`

## Requirements

- Python 3.8+
- `numpy`, `torch>=2.4`, `Pillow`

## Installation

```bash
pip install equiforge
```

## Usage

```python
from equiforge import pers2equi, equi2pers

equi = pers2equi('input.jpg', output_height=1024, fov_x=90.0)

pers = equi2pers(
    'pano.exr',
    output_width=512, output_height=512,
    fov_x=90.0, yaw=45.0,
    use_gpu=True, output_dtype="float32",
)
```

```python
import torch
from equiforge import equi2pers

pano = torch.as_tensor(exr_array, dtype=torch.float32, device="cuda")  # (H,W,3)
for yaw in [0, 90, 180, 270]:
    view = equi2pers(pano, 512, 512, fov_x=90.0, yaw=yaw,
                     output_dtype="float32", device="cuda", as_tensor=True)
```

## API

- `output_dtype="uint8"` (default): clip to `[0,255]`.
- `output_dtype="float32"`: raw samples, no clip.
- `sampling_method="bilinear"` (default) or `"nearest"`.
- `device=None` auto-selects from `use_gpu`; pass `device="cpu"/"cuda"` to override.
- `as_tensor=True` returns a torch tensor instead of numpy.

## Acknowledgements:

- [equilib](https://github.com/haruishi43/equilib)
- [Perspective-and-Equirectangular](https://github.com/timy90022/Perspective-and-Equirectangular)
