Skip to main content

SplatThis

Turn a PNG or JPEG into a portable scene of editable 2D Gaussian splats.

SplatThis fits one Gaussian population and exports it directly to five formats:

Format Artifact
SVG Static, script-free vector gradients
PPTX Native editable PowerPoint shapes, without a bitmap fallback
Canvas Self-contained HTML using the Canvas 2D API
CSS Script-free HTML made from CSS gradients
EML An email message with inline CSS and plain-text/Outlook fallbacks

The default installation is deliberately small: NumPy, Pillow, and Torch. MLX, browser capture, and low-bit population embedding are opt-in.

Install

SplatThis requires Python 3.11 or newer.

pip install splatthis

Optional capabilities are installed separately:

pip install "splatthis[mlx]"      # MLX on Apple Silicon
pip install "splatthis[capture]"  # Playwright client for Chrome/Chromium
pip install "splatthis[steg]"     # low-bit PNG population carrier

Command line

SVG is the default output:

splatthis input.png -o output.svg

Select another target with --format:

splatthis input.png --format pptx   -o output.pptx
splatthis input.png --format canvas -o output-canvas.html
splatthis input.png --format css    -o output-css.html
splatthis input.png --format eml    -o output.eml

Control the fit with a small set of explicit options:

splatthis input.jpg --format pptx \
  --splats 3000 \
  --stages 300,200,100 \
  --max-edge 512 \
  --seed 42 \
  -o output.pptx

Use MLX explicitly on an Apple-Silicon Metal session:

splatthis input.png --optimizer-backend mlx -o output.svg

If MLX cannot allocate a Metal device, SplatThis warns and falls back to Torch before fitting.

Capture native output

Capture SVG, CSS, or Canvas output with an installed Chrome/Chromium browser:

splatthis input.png --format css -o output.html --capture capture.png

On macOS, PPTX capture uses installed Microsoft PowerPoint through OSA and the system screenshot utility, with no additional Python package:

splatthis input.png --format pptx -o output.pptx --capture powerpoint.png

macOS may request Automation, Accessibility, and Screen Recording permission.

Recoverable populations

SVG and PPTX can carry the compressed population that produced them. Preview PNGs can carry the same envelope in metadata and, with the steg extra, in one or two low bits per channel:

splatthis input.png -o output.svg --embed-population
splatthis input.png -o output.svg \
  --preview preview.png \
  --embed-population \
  --embed-population-in-pixels

Embedded populations are derivatives of the input image. Pixel embedding is not a security feature, changes pixels slightly, and has finite capacity.

Python API

from splatthis import SplatConverter

converter = SplatConverter(
    max_splats=2000,
    stages=[200, 150, 100, 50],
    quality_profile="max-fidelity",
    device="cpu",
    seed=42,
)
converter.convert("input.png", "output.pptx", output_format="pptx")

The historical PNG2SVGConverter name remains available. Direct emitters are also public for applications that already have a GaussianSplat population.

Project

SplatThis is released under the MIT License.

Metadata

Release files for splatthis 0.3.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for splatthis 0.3.0
File Size Uploaded Provenance
splatthis-0.3.0.tar.gz 464.1 kB With provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for splatthis 0.3.0
File Interpreter ABI Platform
splatthis-0.3.0-py3-none-any.whl Python 3 none any Details

Total release size: 575.3 kB

Release files / splatthis-0.3.0.tar.gz

Download URL splatthis-0.3.0.tar.gz
Size 464.1 kB
Tags Source
SHA-256 checksum
How to use checksums
7f649c2883f2cf0f66861881888b35627bdfe9014ab90ae31fc31f92004bfb4f
BLAKE2b-256 checksum
How to use checksums
cc48779894f602eb2f7596a7f5beb37c2c1e50cea1409a7072d068043776052b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 7, 2026.

Transparency log

Release files / splatthis-0.3.0-py3-none-any.whl

Download URL splatthis-0.3.0-py3-none-any.whl
Size 111.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
fa88e504c0d8ba4f9099b5b9bcdfed6c91e82c74e877b5aee9d869afd642d29c
BLAKE2b-256 checksum
How to use checksums
e0564f36ee64f74140498f352d4182be5b61f09d1dcc7eb19fbc6eb53d5cf4fb
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 7, 2026.

Transparency log

Security for splatthis 0.3.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for splatthis-0.3.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 7, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page