README.md

metadata

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.