lllyasviel / lllyasviel/stable-diffusion-webui-forge

FaceSwapLab support

Open
#274 13 comments 7 reactions 0 assignees View on GitHub
enhancement
Dominant language
Python
Stars
13k
Forks
1.7k
PR merge metrics
No merged PRs in 30d

Description

### Checklist

- [X] The issue exists after disabling all extensions
- [X] The issue exists on a clean installation of webui
- [X] The issue is caused by an extension, but I believe it is caused by a bug in the webui
- [X] The issue exists in the current version of the webui
- [X] The issue has not been reported before recently
- [ ] The issue has been reported before but has not been fixed yet

### What happened?

I have installed a brand new forge webui from scratch to verify this issue. Basically the extension does not work in the image2image tab specifically. It works in the text2image tab though. This issue is not present in the regular Automatic1111.

### Steps to reproduce the problem

Download faceswaplab extension and attempt to replace a face in the image2image tab.

### What should have happened?

Face is replaced with another face.

### What browsers do you use to access the UI ?

Google Chrome

### Sysinfo

Windows 11, RTX4090, Brand new forge ui installed from scratch with one click installer.

### Console logs

```Shell
100%|████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:00<00:00, 5.43it/s]
2024-02-15 20:12:42,667 - FaceSwapLab - INFO - Try to use model : G:\Forge\webui_forge_cu121_torch21\webui\models\faceswaplab\inswapper_128.onnx
2024-02-15 20:12:42,750 - FaceSwapLab - INFO - Load analysis model det_size=(320, 320), det_thresh=0.5, providers = ['CUDAExecutionProvider'], will take some time. (> 30s)
Loading (320, 320) analysis model (first time is slow): 0%| | 0/1 [00:00

Contributor guide

No contributing guide indexed for this repository

Research direction

Start at scripts/faceswaplab.py line 201 and scripts/faceswaplab_utils/imgutils.py line 224, where the image2image postprocess calls processing.apply_overlay with four arguments. Reproduce the FaceSwapLab face replacement in image2image and inspect the apply_overlay signature. Done means the face is replaced without the reported TypeError.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
backend
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.