Metadata-Version: 2.4
Name: fourdgs
Version: 0.6.0
Summary: Reader and writer for the 4dgs container format for 4D gaussian splat scenes
Author: Avala AI
License-Expression: Apache-2.0
Project-URL: Homepage, https://4dgs.dev
Project-URL: Repository, https://github.com/avala-ai/4dgs
Project-URL: Specification, https://4dgs.dev/docs/spec/
Keywords: gaussian-splatting,4d,container-format,volumetric-video
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.24
Provides-Extra: zstd
Requires-Dist: zstandard>=0.22; extra == "zstd"
Provides-Extra: usd
Requires-Dist: usd-core>=26.3; extra == "usd"
Provides-Extra: dev
Requires-Dist: pytest>=8; extra == "dev"
Requires-Dist: ruff==0.16.0; extra == "dev"
Dynamic: license-file

# 4dgs — Python

Reference implementation of the [4dgs container format](https://github.com/avala-ai/4dgs).

```
pip install fourdgs
```

Reads and writes `.4dgs` files, converts sequences of gaussian splat PLY frames, and inspects and
validates existing files. See the [specification](../website/docs/spec/index.md) for the format and
the [feature matrix](../website/docs/reference/index.md) for what is implemented.

Audio is a collection of independently timed sources with fixed or keyframed scene-space poses. The
SDK reconstructs source state at time `t`; listener pose, HRTF/panning, attenuation and mixing
belong to the player. Indexed callers can read descriptors and source state without fetching payload
bytes, then request bounded encoded ranges per source.
