Skip to content

README.md

Latest commit

 

History

History
120 lines (84 loc) · 4.13 KB

README.md

File metadata and controls

120 lines (84 loc) · 4.13 KB

简体中文 | English

🎥 KLing-Video-WatermarkRemover-Enhancer

GitHub stars GitHub forks GitHub issues

One-click watermark removal for KLing videos - Make your videos more professional! 🚀

Demo

✨ Key Features

🎯 Perfect Watermark Removal

  • Smart detection and precise removal of KLing watermarks
  • Lossless quality with smooth, natural edges
  • Batch processing support for maximum efficiency

🎨 AI Video Enhancement

  • Super-resolution technology powered by Real-ESRGAN
  • Smart optimization of brightness, contrast, and clarity
  • Special facial detail enhancement for better-looking characters

⚡ Efficient & Convenient

  • Simple command-line operation
  • Support for batch video processing
  • Customizable processing parameters

🔧 Installation

Follow these steps to install KLing-Video-WatermarkRemover-Enhancer:

git clone --recursive https://github.com/chenwr727/KLing-Video-WatermarkRemover-Enhancer.git
conda create -n kling python=3.10
conda activate kling
pip install -r requirements.txt

🛠️ Configuration

The config.yaml file defines parameters for watermark removal and video enhancement.

Watermark Removal

Parameters for watermark removal:

  • position: [556, 1233, 701, 1267]

    • Defines the watermark area coordinates
    • Format: (x1, y1, x2, y2), representing top-left (556, 1233) and bottom-right (701, 1267) coordinates
    • Used to determine the watermark region during processing

  • ckpt_p: "./weights/sttn.pth"

    • Path to the watermark removal model weights
    • sttn.pth is based on STTN (Spatio-Temporal Trajectory Network)

  • mask_expand: 30

    • Pixel expansion of the watermark area
    • Ensures clean removal without artifacts

  • neighbor_stride: 10

    • Stride size for temporal neighborhood computation
    • Controls sampling frequency of neighboring frames

Video Enhancement

Parameters for video enhancement:

  • RealESRGAN_model_path: "./weights/RealESRGAN_x2plus.pth"

    • Path to Real-ESRGAN model weights for video quality enhancement
    • More models available in Real-ESRGAN/inference_realesrgan.py
    • RealESRGAN_x2plus.pth

  • GFPGANer_model_path: "./weights/GFPGANv1.4.pth"

    • Path to GFPGAN model weights for facial enhancement
    • More models available in Real-ESRGAN/cog_predict.py
    • GFPGANv1.4.pth

🚀 Quick Start

Start enhancing your videos with just one command:

python main.py --input your_video.mp4 --remove-watermark --enhance-video

🎮 Common Usage Examples

# Remove watermark only
python main.py --input video.mp4 --remove-watermark

# Enhance video only
python main.py --input video.mp4 --enhance-video

# Process all videos in a folder
python main.py --input videos_folder --remove-watermark --enhance-video

📁 Project Structure

KLing-Video-WatermarkRemover-Enhancer/
├── 📄 main.py          # Main program entry
├── 📁 modules/         # Core function modules
├── 📁 utils/          # Utility functions
└── 📄 config.yaml     # Configuration file

🤝 References

  • STTN - Powerful video inpainting technology
  • Real-ESRGAN - State-of-the-art video super-resolution solution

🌟 Support

If this project helps you, please consider giving it a star ⭐️