silukman_video_enhancer
silukman_video_enhancer is a local-first, offline video enhancement application designed to run entirely on your own machine. It improves video quality without sending data to the cloud, utilizing hardware-accelerated AI models locally.
This project is a video-focused evolution of silukman_image_enhancer.
Repository social preview asset: docs/assets/github-social-preview.svg.
Key Features
- Offline Video Upscaling: Increase video resolution (e.g., 720p to 1080p, 1080p to 4K) using local neural networks.
- Denoising & Deblurring: Remove camera noise, low-light grain, and light motion blurs.
- Compression Artifact Cleanup: Clean up pixelation and blocky artifacts caused by aggressive video compression.
- Color Correction: Simple automatic adjustments to restore vibrance, contrast, and color balance.
- Audio Preservation: Automatically extracts, syncs, and merges the original audio back into the enhanced video.
- Hardware Acceleration: Built on ONNX Runtime, leveraging local GPU/NPU acceleration (NVIDIA CUDA, Apple CoreML, Windows DirectML) where available.
- CLI Workflow: Optimized command-line interface for batch processing and integration into automation scripts.
- Python Desktop UI: Planned cross-platform desktop interface using a Python GUI stack so the CLI, pipeline, inference, and UI layers can share one language/runtime.
Technical Architecture (Simplified)
[Input Video] ──> [FFmpeg Frame Extraction] ──> [Frame Queue]
│
▼
[Output Video] <── [FFmpeg Muxing + Audio] <── [ONNX Inference Engine]
Prerequisites
To run this application locally, you will need:
- Python 3.9+
- FFmpeg installed and added to your system's PATH (if missing, the application will automatically download static binaries to a local
bin/folder on first run). - Hardware Accelerator Drivers (optional but highly recommended):
- NVIDIA CUDA Toolkit (for NVIDIA GPUs)
- DirectML (supported out-of-the-box on Windows)
- CoreML (supported out-of-the-box on macOS)
- Desktop UI Toolkit (planned): PySide6/Qt for the Python desktop application.
Installation
# Clone the repository
git clone https://github.com/lukman-ss/silukman_video_enhancer.git
cd silukman_video_enhancer
# Set up virtual environment
python3 -m venv .venv
source .venv/bin/activate # On Windows use: .venv\Scripts\activate
# Upgrade pip (required for pyproject.toml editable installs)
pip install --upgrade pip
# Option 1: Install directly from PyPI (recommended)
pip install silukman-video-enhancer
# Option 2: Install in development/editable mode from source
pip install --no-compile -r requirements.txt -e ".[onnx,dev]"
# Option 3: Install a wheel downloaded from the tagged GitHub Release assets
pip install silukman_video_enhancer-*.whl
Usage
# Basic enhancement (upscaling)
python -m app enhance -i input.mp4 --scale 2 --output output.mp4
# Run denoise and color correction
python -m app enhance -i input.mp4 --denoise --color-correct --output output.mp4
# Launch the Python desktop app
python -m app desktop
# or
python -m ui.desktop
Project Status & Roadmap
This project is currently in active implementation. Phase 1 through Phase 10 are documented as completed.
- Phase 1 (MVP): Core video pipeline with frame-by-frame extraction, ONNX scaling model integration, and CLI support.
- Phase 2 (Refinement): Temporal consistency checks, advanced noise/blur removal models, and batch CLI processing.
- Phase 3 (Desktop UI): Python desktop interface with PySide6/Qt, side-by-side preview, and offline packaging.
- Phase 4 (Production): Release QA, job control, render farm hardening, subtitle engines, INT8, and timeline preview.
- Phase 5 (Advanced Media & Runtime): Vulkan/WebGPU, OpenVINO/QNN, HDR/LUT workflows, tone mapping, professional codecs, hardware encoder profiling, high-resolution render planning, and model toolchain improvements.
- Phase 6 (Headless API & Operations): REST API, client tooling, access controls, durable queues, worker pools, retries, graceful shutdown, LAN discovery, service diagnostics, log streaming, disk quotas, observability, and containerized render nodes.
- Phase 7 (Ecosystem Governance & Lifecycle): Plugin SDK, sandboxing, workflow automation, reproducible manifests, configuration migration, audit logs, compatibility matrix tests, signed updates, model quarantine, telemetry, maintenance, and SDK docs.
- Phase 8 (Desktop UX & Batch Processing): Multi-file queueing, batch progress, cancellation, retry, recent files, ETA, post-job actions, queue reordering, and output format selection.
- Phase 9 (CI/CD & Cross-Platform Release Workflow): GitHub Actions test matrix, version tagging, signed Windows/macOS/Linux release artifacts, draft releases, smoke tests, and dependency caching.
- Phase 10 (GitHub Repository Presence): Repository metadata, social preview asset, release badges, release/package publication workflows, issue templates, release naming policy, and security disclosure policy.
GitHub Repository Setup
Repository metadata is documented in .github/repository.yml. Apply the same values in GitHub repository settings:
- Description:
Local-first offline video enhancement CLI and Python desktop app. - Website:
https://github.com/lukman-ss/silukman_video_enhancer - Topics:
video-enhancement,upscaling,onnx,ffmpeg,pyside6,ai,python,offline-first,desktop-app - Social preview: upload docs/assets/github-social-preview.svg
Contributing
Contributions are welcome! Please read our Documentation first to understand the architecture and pipeline.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Metadata
Release files for silukman-video-enhancer 0.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| silukman_video_enhancer-0.2.0.tar.gz | 133.9 kB | Without provenance | Details for silukman_video_enhancer-0.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| silukman_video_enhancer-0.2.0-py3-none-any.whl | Python 3 | none | any | Details for silukman_video_enhancer-0.2.0-py3-none-any.whl |
Total release size: 254.6 kB
Release files / silukman_video_enhancer-0.2.0.tar.gz
| Download URL | silukman_video_enhancer-0.2.0.tar.gz |
|---|---|
| Size | 133.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
66f5111b16db2a22191a59e0487565f18494ed0a7b9b03553a16004db443fd7c
|
|
BLAKE2b-256 checksum How to use checksums |
c1fab72097aa63194d72aacd7b5387b69e7a84873da1ee4668976e6428abda1b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Release files / silukman_video_enhancer-0.2.0-py3-none-any.whl
| Download URL | silukman_video_enhancer-0.2.0-py3-none-any.whl |
|---|---|
| Size | 120.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
8254f106e98beaec7bf60c7bf55c8877bf8dee53cc81821ebdb813d4cfb73e15
|
|
BLAKE2b-256 checksum How to use checksums |
f4b55ac12c4585c47da3caaba96d5e7d3509818077ce2c9b0470070d7bef9af6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Security for silukman-video-enhancer 0.2.0
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
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.
To learn more about protecting yourself from malicious actors, visit our package security documentation.