hxngiee / diffface Goto Github PK
View Code? Open in Web Editor NEWDiffFace: Diffusion-based Face Swapping with Facial Guidance
License: Other
DiffFace: Diffusion-based Face Swapping with Facial Guidance
License: Other
Traceback (most recent call last):
File "/data/docker/yerang/DiffFace/main.py", line 6, in
image_editor = ImageEditor(args)
File "/data/docker/yerang/DiffFace/optimization/image_editor.py", line 123, in init
self.fa = face_alignment.FaceAlignment(face_alignment.LandmarksType._2D, flip_input=False)
File "/opt/conda/lib/python3.9/enum.py", line 429, in getattr
raise AttributeError(name) from None
AttributeError: _2D
Solution:
Replace image_editor.py file
"_2D" -> "TWO_D"
It works!!!!!! 👍👍🔥
Does the current released code version support DDIM sampling? I have tried to tag the "--ddim", but got weird results
Failure to disclose the checkpoint model disproves a false thesis.
Thanks for your work on face swapping using diffusion models. When do you plan to publish the code?
Best Regards
Thanks for releasing the code & models.
I'm getting cuda memory error on 8gb card. What is the expected Vram?
Setting device to 'cpu' causes error.
Hello, author, thank you for sharing. I have a question: Do you have a specific method for aligning the faces of the source image and the target image? Looking forward to your early reply. Thanks!
When will the code be open source?
Should we perform element-wise addition of id vector and Timestep t?
Thank you!!!
Hi,
Can you please share the code of the repo? If that's not possible, please share the trained models for inference.
I really appreciate any help you can provide.
I modify the parsing index with [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,17],including 17 hair, however, the results seems no diff, can you help me with that?
Is your model brings pretraied DDPM(trained with FFHQ) and finetune with loss? In your code, it seems like you finetune the model per each image using Model.pt, but there is no word about finetune in your paper. Am i right?
Or, is it updating noise like using classifier guidance?
Thank you for your nice work, and it helped me a lot.
DiffFace output is very good.
Please release the code and models of DiffFace.
Thanks
Thank you for your contribution! Please release the code and models of DiffFace.
Traceback (most recent call last):
File "main.py", line 7, in
image_editor.edit_image_by_prompt()
File "/root/DiffFace/optimization/image_editor.py", line 393, in edit_image_by_prompt
visualization_path = visualization_path.with_stem(f"{visualization_path.stem}i{iteration_number}b{0}")
AttributeError: 'PosixPath' object has no attribute 'with_stem'
Dear authors,
Thanks for releasing this project, I do admire the work you guys have done!
I kindly ask you to release the training code, which is important for the community to test ideas and thoughts.
(Even a minimal scripts for training is adequate.)
Best regards!
File "/workspace/user_code/swapface_pro/DiffFace/main.py", line 6, in
image_editor = ImageEditor(args)
File "/workspace/user_code/swapface_pro/DiffFace/optimization/image_editor.py", line 123, in init
self.fa = face_alignment.FaceAlignment(face_alignment.LandmarksType.TWO_D, flip_input=False)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/site-packages/face_alignment/api.py", line 77, in init
self.face_detector = face_detector_module.FaceDetector(device=device, verbose=verbose, **face_detector_kwargs)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/site-packages/face_alignment/detection/sfd/sfd_detector.py", line 24, in init
model_weights = load_url(models_urls['s3fd'])
File "/data/miniconda3/envs/DiffFace/lib/python3.9/site-packages/torch/hub.py", line 586, in load_state_dict_from_url
download_url_to_file(url, cached_file, hash_prefix, progress=progress)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/site-packages/torch/hub.py", line 452, in download_url_to_file
u = urlopen(req)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 214, in urlopen
return opener.open(url, data, timeout)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 517, in open
response = self._open(req, data)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 534, in _open
result = self._call_chain(self.handle_open, protocol, protocol +
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 494, in _call_chain
result = func(*args)
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 1389, in https_open
return self.do_open(http.client.HTTPSConnection, req,
File "/data/miniconda3/envs/DiffFace/lib/python3.9/urllib/request.py", line 1349, in do_open
raise URLError(err)
urllib.error.URLError: <urlopen error Tunnel connection failed: 403 Forbidden>
hi, i have this issue,how can i resolve it
Hey, thanks for your excellent work, i have some question about Facial Guidance Module.
Have you retrain the model(identity guidance mdoel、semantic guidance model、gaze guidance model) on noisy images Xt, or you just using the pretained model in original image X0 for guidance.
Looking forward to your reply.Thanks.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.