Spaces:
Running
Running
File size: 2,105 Bytes
61cde45 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
---
title: 'FaceFusion Pro - AI Face Swap & Enhancement (v3.3.2)'
emoji: 🎭
colorFrom: red
colorTo: pink
sdk: gradio
sdk_version: 5.39.0
app_file: app.py
pinned: false
short_description: Professional AI face swapping with enhanced UI
---
# FaceFusion - Advanced Face Swapping Tool
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
## Optimal Settings for Best Quality Face Swapping
This FaceFusion setup has been pre-configured with optimal settings for the highest quality output:
### Default Processors
- **Face Swapper**: Enabled by default with `inswapper_128` model
- **Face Enhancer**: Enabled to improve final quality with `gfpgan_1.4` model
### Face Swapper Configuration
- **Model**: `inswapper_128` - One of the best quality models available
- **Pixel Boost**: `1024x1024` - Highest resolution for maximum detail
- **Alternative**: You can also try `blendswap_256` model for different results
### Face Enhancer Configuration
- **Model**: `gfpgan_1.4` - High-quality face restoration
- **Blend**: 80% - Balances enhancement with natural appearance
- **Alternative Models**: Try `restoreformer_plus_plus` for different enhancement styles
### Output Quality
- **Image Quality**: 95 (maximum quality with minimal compression)
- **Video Quality**: 95 (maximum quality)
- **Video Preset**: slow (best compression efficiency, higher quality)
### Usage Tips
1. **Source**: Upload the face image you want to apply
2. **Target**: Upload the image/video where you want to swap the face
3. **Processors**: Keep both face_swapper and face_enhancer selected for best results
4. **Face Selector**: Use "Reference" mode to track specific faces in videos
5. **Execution Providers**: Select CUDAExecutionProvider if you have NVIDIA GPU for faster processing
### Performance Notes
- Higher quality settings require more processing time and memory
- For faster processing, you can reduce pixel boost to 512x512 or 768x768
- Video memory strategy can be adjusted based on your GPU VRAM
All settings can be modified in the UI or by editing `facefusion.ini`. |