|
Download README.md from maximsobolev275/BFS-Qwen21-Best-Face-Swap-Fix: direct link, hf CLI and curl.
- Browser
- Download file 2.54 kB
-
https://huggingface.co/maximsobolev275/BFS-Qwen21-Best-Face-Swap-Fix/resolve/main/README.md
- Command line
-
hf download hf://maximsobolev275/BFS-Qwen21-Best-Face-Swap-Fix/README.md
-
curl -L -o README.md https://huggingface.co/maximsobolev275/BFS-Qwen21-Best-Face-Swap-Fix/resolve/main/README.md
2.54 kB
| license: mit | |
| base_model: Alissonerdx/BFS-Best-Face-Swap | |
| tags: | |
| - lora | |
| - face-swap | |
| - comfyui | |
| - dare | |
| - pruning | |
| # BFS (Best Face Swap) β Optimized & De-blurred LoRA Variants | |
| This repository contains cleaned and optimized variations of the [Alissonerdx/BFS-Best-Face-Swap](https://huggingface.co/Alissonerdx/BFS-Best-Face-Swap) LoRA. | |
| ### Motivation & Problem | |
| While the original BFS LoRA provides solid face swap transfer, it often introduces significant **blurriness, fine texture loss, and a plastic/soapy degradation** across skin areas. | |
| To resolve this issue while retaining identity likeness, these variants were optimized using weight pruning and stochastic drop masking: | |
| - **Pruning (20%)**: Removes the bottom 20% weakest/noisy weight layers that cause unwanted model drift and texture degradation. | |
| - **DARE Drop Masking (without Rescaling)**: Random drop-masking (density 0.50β0.60) applied **without** magnitude rescaling. Skipping the rescaling step prevents artificial weight amplification, successfully stripping away soapy blur while keeping the facial structure intact. | |
| --- | |
| ### Available Variants | |
| | File | Pruning & Masking | Description | | |
| | :--- | :--- | :--- | | |
| | `bfs_dare60-avgrank30.safetensors` | Drop Mask (density 0.60, no rescale), avg rank 30 | Gentle cleanup; preserves strong likeness while noticeably reducing original blur. | | |
| | `bfs_prune20_dare50-avgrank26.safetensors` | 20% Pruned + Drop Mask (density 0.50, no rescale), avg rank 26 | Strongest de-blurring and detail retention; best for crisp skin textures and sharp eyes. | | |
| | `bfs_prune20_dare60-avgrank30.safetensors` | 20% Pruned + Drop Mask (density 0.60, no rescale), avg rank 30 | Balanced profile: removes muddy blur while keeping likeness very close to the original. | | |
| --- | |
| ## Reference Inputs | |
| | Reference Head (`ref_head.png`) | Reference Body (`ref_body.png`) | | |
| | :---: | :---: | | |
| | <img src="ref_head.png" width="400" /> | <img src="ref_body.png" width="400" /> | | |
| --- | |
| ## Visual Comparisons Across Seeds | |
| The comparison grids below demonstrate: | |
| 1. **Base Model** (No LoRA applied) | |
| 2. **Original BFS LoRA** (Notice the blurriness and smoothed skin texture) | |
| 3. **Optimized Variants** (Pruned / DARE drop-masked versions restoring sharp details) | |
| ### Comparison β Seed 0 | |
|  | |
| ### Comparison β Seed 1 | |
|  | |
| ### Comparison β Seed 2 | |
|  | |
| ### Comparison β Seed 3 | |
|  | |