e4s2022 / e4s Goto Github PK

View Code? Open in Web Editor NEW

356.0 356.0 31.0 82.07 MB

(CVPR 2023) E4S: Fine-grained Face Swapping via Regional GAN Inversion

Home Page: https://e4s2022.github.io

License: MIT License

Shell 0.08% Python 95.14% C++ 0.63% Cuda 4.15%

face-editing face-swapping gan

e4s's People

Contributors

Stargazers

Watchers

e4s's Issues

Where can I find the demo of the video face swap

The second pair image will fail, do we need clean some cache?

how to remove source image neck from result image?

how to remove source image neck from result image?
Thanks

The video swap is not consistent

Thanks for your excellent work. I ran face_swap for each frame in the video but found the results were not smooth.
But in your video demo, the results are very nice, could you please tell me how to fix it?
Thanks for your help!

Quality improvement - Hard Edges for Neck - Where to adjust the lambda value?

Hi, I see in my results that neck has hard edges. Same as in you shared image in this issue
Is there a way to solve for that? I also saw the youtube video, where a lambda variable is discussed as a controllable face swap attribute. How can one change that in the face_swap script?

Model license

Hello!

What is a model license?

Can I use it for commercial purpose?

Thanks!

requirements.txt

Provide a way to do this without miniconda / anaconda.

Keeping images of source

How do you suggest modifying the code such that it keeps the expressions of the source image?

while running gradio_demo.py getting Segmentation fault (core dumped)

resolve this is the architecture error? @e4s2022

docker image

would anyone be so kind to make a docker image? unfortunately this is what I always end up resorting to, as for some reason everything crashes on both windows and ubuntu (wsl2). ive been trying to use this software for more than a week now, installing multiple torch/cudatoolkit versions to no avail.

Admin of the project, if youre reading this, please send me a message on telegram https://t.me/keronikateka

How to use it on my own images

Thank you for sharing such a wonderful project, but when I use FFHQ pre-trained weights to swap my own data, the result is very bad, please tell me how to do it if I want to swap my own images.

bad results on unseen data

Thanks for your work!
I wonder have you ever tried testing on the images that are not sampled from FFHQ? I test it on my daily images, and it returns bad results. (aligned faces are used and try on different resolution)

Error running example image

'''
Traceback (most recent call last):
File "e4s/scripts/face_swap.py", line 392, in
faceSwapping_pipeline(opts.source, opts.target, opts, save_dir=opts.output_dir,
File "python3.9/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "e4s/scripts/face_swap.py", line 314, in faceSwapping_pipeline
swapped_and_pasted = smooth_face_boundry(swapped_face_image, T, full_mask_image, radius=outer_dilation)
File "e4s/scripts/face_swap.py", line 95, in smooth_face_boundry
image_masked.putalpha(blurred_mask)
File "python3.9/site-packages/PIL/Image.py", line 1705, in putalpha
self.im.putband(alpha.im, band)
ValueError: images do not match
'''

Please assist. Thank you!

Error: pretrained model

Hi, when I tried swap face operation:

python scripts/face_swap.py --source=example/input/faceswap/source.jpg --target=example/input/faceswap/target.jpg

the following error is reported：

Traceback (most recent call last):
  File "scripts/face_swap.py", line 345, in <module>
    generator, kp_detector, he_estimator, estimate_jacobian = init_facevid2vid_pretrained_model(face_vid2vid_cfg, face_vid2vid_ckpt)
  File "/root/project/e4s-main/./src/pretrained/face_vid2vid/driven_demo.py", line 216, in init_facevid2vid_pretrained_model
    generator, kp_detector, he_estimator = load_checkpoints(config_path=cfg_path, checkpoint_path=ckpt_path, gen="spade", cpu=False)
  File "/root/project/e4s-main/./src/pretrained/face_vid2vid/driven_demo.py", line 49, in load_checkpoints
    checkpoint = torch.load(checkpoint_path)
  File "/root/miniconda3/envs/e4s/lib/python3.8/site-packages/torch/serialization.py", line 712, in load
    return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
  File "/root/miniconda3/envs/e4s/lib/python3.8/site-packages/torch/serialization.py", line 1049, in _load
    result = unpickler.load()
  File "/root/miniconda3/envs/e4s/lib/python3.8/site-packages/torch/serialization.py", line 1019, in persistent_load
    load_tensor(dtype, nbytes, key, _maybe_decode_ascii(location))
  File "/root/miniconda3/envs/e4s/lib/python3.8/site-packages/torch/serialization.py", line 997, in load_tensor
    storage = zip_file.get_storage_from_record(name, numel, torch._UntypedStorage).storage()._untyped()
RuntimeError: PytorchStreamReader failed reading file data/94709563812016: invalid header or archive is corrupted

It seems to be related to the pre-trained model 00000189-checkpoint.pth.tar, but I downloaded it from the given link.
Can someone with a similar problem help me out?
Thanks.

The inference result is not good

The inference result is not good, is this a model that only works well for trained faces?

Result
source, target, result

Furthermore, The results you posted are also different from the result image of the paper.
Result
source, target, my result

Here is your result on paper

Please provide detailed answers on the above results.

How does this lib compare to insightface for face swap?

https://github.com/deepinsight/insightface tested has been very accurate without any training. I haven't been able to run this lib successfully but curious if it produces better results than insightface without any training.

checkpoint gone

hi , the weight link of 00000189-checkpoint.pth.tar in the facevid2vid paper is gone,can you send me a copy one？ thank you very much! my email is [email protected]

RuntimeError: shape '[-1, 512, 8, 8]' is invalid for input of size 41472. error in CPU environment.

(e4s_env) [root@www e4s]# python face_swap.py --source=example/input/faceswap/source.jpg --target=example/input/faceswap/target.jpg --device=cpu --checkpoint_path=/root/e4s/pretrained_ckpts/e4s/iteration_300000.pt --output_dir=/root/temp/ --verbose=True
Load face_vid2vid pre-trained model success!
Load GPEN pre-traiend model success!
Load pre-trained face parsing models success!
Load E4S pre-trained model success!
Traceback (most recent call last):
File "/root/e4s/face_swap.py", line 388, in
faceSwapping_pipeline(opts.source, opts.target, opts, save_dir=opts.output_dir,
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/root/e4s/face_swap.py", line 242, in faceSwapping_pipeline
driven_face, _ , structure_feats = net.gen_img(torch.zeros(1,512,32,32).to(opts.device), driven_style_codes, driven_onehot)
File "/root/e4s/src/models/networks.py", line 175, in gen_img
images, result_latent, structure_feats = self.G([style_codes], struc_codes, mask, input_is_latent=True,
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/root/e4s/src/models/stylegan2/model.py", line 640, in forward
out = conv1(out, latent[:, :, i], mask, noise=noise1)
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/root/e4s/src/models/stylegan2/model.py", line 396, in forward
out_i = self.conv(input, style[:, comp_i, :])
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/root/e4s/src/models/stylegan2/model.py", line 300, in forward
out = self.blur(out)
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/root/e4s/src/models/stylegan2/model.py", line 92, in forward
out = upfirdn2d(input, self.kernel, pad=self.pad)
File "/root/e4s/src/models/stylegan2/op/upfirdn2d.py", line 142, in upfirdn2d
out = UpFirDn2d.apply(
File "/root/e4s/e4s_env/lib64/python3.9/site-packages/torch/autograd/function.py", line 506, in apply
return super().apply(*args, **kwargs) # type: ignore[misc]
File "/root/e4s/src/models/stylegan2/op/upfirdn2d.py", line 118, in forward
out = out.view(-1, channel, out_h, out_w)
RuntimeError: shape '[-1, 512, 8, 8]' is invalid for input of size 41472

please suggest.
Thanks

faceswap的结果不成人样，可能是什么原因？

用项目自带的图片测试没问题，但是用网上的图片测试，生成的结果惨不忍睹，都不敢传上来，怕吓着人。

Google Colab

Can someone create a google colab for this?

checkpoint

The training process encountered No such file or directory: './pretrained_ckpts/auxiliray/model.pth' This checkpoint does not seem to be mentioned?

Where we can control the lambda parameter ?

Hi @e4s2022 , How can i control the percentage of the swapping ?

Can it be run on Windows?

I hope to be able to run it on Windows. I haven't tried it yet and I'm not sure if it can be successful.

Does the model work with PyTorch v2.x?

Hi, I was trying to run the demo with Pytorch v2.x. It seems that the swapped face is black.
When i switched the platform back to v1.x, it works.
Anyone had the same problem? Any suggestion to make this work with PT v2.x?

Thanks!

Can we run face swap with CPU and without GPU?

Hello,
I got this error : when I tried face swap.

Traceback (most recent call last):
File "/usr/lib64/python3.9/runpy.py", line 188, in _run_module_as_main
mod_name, mod_spec, code = _get_module_details(mod_name, _Error)
File "/usr/lib64/python3.9/runpy.py", line 111, in _get_module_details
import(pkg_name)
File "/root/e4s/scripts/face_swap.py", line 17, in
from src.pretrained.face_parsing.face_parsing_demo import init_faceParsing_p retrained_model, faceParsing_demo, vis_parsing_maps
File "/root/e4s/src/pretrained/face_parsing/face_parsing_demo.py", line 13, in
from src.pretrained.face_parsing.model import BiSeNet, seg_mean, seg_std
File "/root/e4s/src/pretrained/face_parsing/model.py", line 15, in
seg_mean = torch.from_numpy(np.array([[0.485, 0.456, 0.406]])).float().cuda( ).reshape(1,3,1,1)
File "/usr/local/lib64/python3.9/site-packages/torch/cuda/init.py", line 2 47, in _lazy_init
torch._C._cuda_init()
RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.a spx

Please suggest if this can be used with CPU instead of GPU.
Thanks.

AttributeError: 'Editor' object has no attribute 'opt'

(e4s) root@autodl-container-fe7811a13c-a73d6848:~/autodl-tmp/FinegrainedFace/e4s# python face_edit.py --source=./example/input/faceedit/source.jpg --reference=./example/input/faceedit/reference.jpg --region hair eyes --alpha=1
Traceback (most recent call last):
File "face_edit.py", line 107, in
editor = Editor(opts)
File "face_edit.py", line 34, in init
self.faceParsing_model = init_faceParsing_pretrained_model(self.opts.faceParser_name, self.opt.faceParsing_ckpt, self.opt.segnext_config)
AttributeError: 'Editor' object has no attribute 'opt'

help v2

this time i tried in ubuntu using wsl2 and conda

(e4s) user@LAPTOP-E4D0U85P:~/e4s$ python scripts/face_swap.py --source=example/input/faceswap/source.jpg --target=example/input/faceswap/target.jpg
Traceback (most recent call last):
File "/home/user/.local/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1516, in _run_ninja_build
subprocess.run(
File "/home/user/miniconda3/envs/e4s/lib/python3.8/subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "scripts/face_swap.py", line 16, in
from src.pretrained.gpen.gpen_demo import init_gpen_pretrained_model, GPEN_demo
File "/home/user/e4s/scripts/src/pretrained/gpen/gpen_demo.py", line 15, in
from src.pretrained.gpen.face_enhancement import FaceEnhancement
File "/home/user/e4s/scripts/src/pretrained/gpen/face_enhancement.py", line 11, in
from src.pretrained.gpen.face_model.face_gan import FaceGAN
File "/home/user/e4s/scripts/src/pretrained/gpen/face_model/face_gan.py", line 14, in
from src.pretrained.gpen.face_model.gpen_model import FullGenerator, FullGenerator_SR
File "/home/user/e4s/scripts/src/pretrained/gpen/face_model/gpen_model.py", line 16, in
from src.pretrained.gpen.face_model.op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d
File "/home/user/e4s/scripts/src/pretrained/gpen/face_model/op/init.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "/home/user/e4s/scripts/src/pretrained/gpen/face_model/op/fused_act.py", line 13, in
fused = load(
File "/home/user/.local/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 969, in load
return jit_compile(
File "/home/user/.local/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1176, in jit_compile
write_ninja_file_and_build_library(
File "/home/user/.local/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1280, in write_ninja_file_and_build_library
run_ninja_build(
File "/home/user/.local/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1538, in run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'fused': [1/2] /usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -isystem /home/user/.local/lib/python3.8/site-packages/torch/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/TH -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/THC -isystem /home/user/miniconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS --expt-relaxed-constexpr -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/user/e4s/scripts/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
FAILED: fused_bias_act_kernel.cuda.o
/usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -isystem /home/user/.local/lib/python3.8/site-packages/torch/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/TH -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/THC -isystem /home/user/miniconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/user/e4s/scripts/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
/usr/include/c++/11/bits/std_function.h:435:145: error: parameter packs not expanded with ‘...’:
435 | function(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:435:145: note: ‘_ArgTypes’
/usr/include/c++/11/bits/std_function.h:530:146: error: parameter packs not expanded with ‘...’:
530 | operator=(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:530:146: note: ‘_ArgTypes’
ninja: build stopped: subcommand failed.

(e4s) user@LAPTOP-E4D0U85P:~/e4s$ conda list

packages in environment at /home/user/miniconda3/envs/e4s:

Getting Error: Maybe a CUDA Version + GCC Version Conflict?

What CUDA version and GCC version should I be using?

Testing inside Ubuntu latest with CUda 11.5 and GCC 11.3

Getting the following.

`raceback (most recent call last):
File "/home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1808, in _run_ninja_build
subprocess.run(
File "/home/g/anaconda3/envs/e4s/lib/python3.8/subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "scripts/face_swap.py", line 16, in
from src.pretrained.gpen.gpen_demo import init_gpen_pretrained_model, GPEN_demo
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/gpen_demo.py", line 15, in
from src.pretrained.gpen.face_enhancement import FaceEnhancement
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_enhancement.py", line 11, in
from src.pretrained.gpen.face_model.face_gan import FaceGAN
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/face_gan.py", line 14, in
from src.pretrained.gpen.face_model.gpen_model import FullGenerator, FullGenerator_SR
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/gpen_model.py", line 16, in
from src.pretrained.gpen.face_model.op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/op/init.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "/media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/op/fused_act.py", line 13, in
fused = load(
File "/home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1202, in load
return jit_compile(
File "/home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1425, in jit_compile
write_ninja_file_and_build_library(
File "/home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1537, in write_ninja_file_and_build_library
run_ninja_build(
File "/home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1824, in run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'fused': [1/2] /usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/TH -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/THC -isystem /home/g/anaconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
FAILED: fused_bias_act_kernel.cuda.o
/usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/TH -isystem /home/g/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/THC -isystem /home/g/anaconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /media/g/PCXtend/1TB/AI/e4s/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
/usr/include/c++/11/bits/std_function.h:435:145: error: parameter packs not expanded with ‘...’:
435 | function(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:435:145: note: ‘_ArgTypes’
/usr/include/c++/11/bits/std_function.h:530:146: error: parameter packs not expanded with ‘...’:
530 | operator=(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:530:146: note: ‘_ArgTypes’
ninja: build stopped: subcommand failed.
`

CelebAHQ dataset incompatibility

The dataloader you've provided seem to be incompatible with the CelebAHQ dataset.
CelebAHQ dataset does not provide the segmentation mask as a whole, but rather in categories.
It is also divided into subfolders depending on the numbers of the image, from 0~14

IndexError: The shape of the mask [1, 1, 1, 512, 512] at index 2 does not match the shape of the indexed tensor [1, 1, 512, 512] at index 2

I tried running the code using the image in the sample, but encountered the following error. I am a beginner in this field, how can I solve it?
IndexError: The shape of the mask [1, 1, 1, 512, 512] at index 2 does not match the shape of the indexed tensor [1, 1, 512, 512] at index 2

pythorch cuda problem.

/python3.8/site-packages/torch/cuda/init.py", line 190, in _lazy_init
torch._C._cuda_init()
RuntimeError: cuda runtime error (101) : invalid device ordinal at /pytorch/aten/src/THC/THCGeneral.cpp:47

No moduel named 'src'

(e4s) jack@jack-All-Series:~/e4s$ python scripts/face_swap.py --source=example/input/faceswap/source.jpg --target=example/input/faceswap/target.jpg
Traceback (most recent call last):
  File "scripts/face_swap.py", line 15, in <module>
    from src.pretrained.face_vid2vid.driven_demo import init_facevid2vid_pretrained_model, drive_source_demo
ModuleNotFoundError: No module named 'src'
(e4s) jack@jack-All-Series:~/e4s$

for context I didn't install any of the optional repos but have downloaded the pre-trained facevid2vid model

customized target_mask.

which script is used for pre-processing to get target_mask image?

Number of iterations for one A100

Due to limited computing resources, I use an A100 to train 300k on FFHQ, batch size = 2. The test effect is not good, how many iterations should be done when there is only one A100?The picture below is the test result：

help :)

I had to move the src folder inside scripts to fix paths

(e4s) C:\Users\user\e4s>python scripts/face_swap.py --source=example/input/faceswap/source.jpg --target=example/input/faceswap/target.jpg
INFO: Could not find files for the given pattern(s).
Traceback (most recent call last):
File "scripts/face_swap.py", line 26, in
from src.models.networks import Net3
File "C:\Users\user\e4s\scripts\src\models\networks.py", line 8, in
from src.models.stylegan2.model import EqualLinear
File "C:\Users\user\e4s\scripts\src\models\stylegan2\model.py", line 11, in
from src.models.stylegan2.op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d, conv2d_gradfix
File "C:\Users\user\e4s\scripts\src\models\stylegan2\op_init_.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "C:\Users\user\e4s\scripts\src\models\stylegan2\op\fused_act.py", line 9, in
fused = load(
File "C:\Users\user\miniconda3\envs\e4s\lib\site-packages\torch\utils\cpp_extension.py", line 1284, in load
return _jit_compile(
File "C:\Users\user\miniconda3\envs\e4s\lib\site-packages\torch\utils\cpp_extension.py", line 1509, in _jit_compile
_write_ninja_file_and_build_library(
File "C:\Users\user\miniconda3\envs\e4s\lib\site-packages\torch\utils\cpp_extension.py", line 1611, in _write_ninja_file_and_build_library
_write_ninja_file_to_build_library(
File "C:\Users\user\miniconda3\envs\e4s\lib\site-packages\torch\utils\cpp_extension.py", line 2048, in _write_ninja_file_to_build_library
_write_ninja_file(
File "C:\Users\user\miniconda3\envs\e4s\lib\site-packages\torch\utils\cpp_extension.py", line 2188, in _write_ninja_file
cl_paths = subprocess.check_output(['where',
File "C:\Users\user\miniconda3\envs\e4s\lib\subprocess.py", line 415, in check_output
return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
File "C:\Users\user\miniconda3\envs\e4s\lib\subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['where', 'cl']' returned non-zero exit status 1.

(e4s) C:\Users\user\e4s>conda list

packages in environment at C:\Users\user\miniconda3\envs\e4s:

Interactive Face Beautification System

Thanks for the great work,

Are there any chance that we can get our hands on Interactive Face Beautification System as in poster presentation

Seems like something wrong with my C++ complier.

ERROR LOG:
ple/input/faceswap/target.jpg
Traceback (most recent call last):
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1808, in _run_ninja_build
subprocess.run(
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "scripts/face_swap.py", line 16, in
from src.pretrained.gpen.gpen_demo import init_gpen_pretrained_model, GPEN_demo
File "/home/hou/code/e4s-main/src/pretrained/gpen/gpen_demo.py", line 15, in
from src.pretrained.gpen.face_enhancement import FaceEnhancement
File "/home/hou/code/e4s-main/src/pretrained/gpen/face_enhancement.py", line 11, in
from src.pretrained.gpen.face_model.face_gan import FaceGAN
File "/home/hou/code/e4s-main/src/pretrained/gpen/face_model/face_gan.py", line 14, in
from src.pretrained.gpen.face_model.gpen_model import FullGenerator, FullGenerator_SR
File "/home/hou/code/e4s-main/src/pretrained/gpen/face_model/gpen_model.py", line 16, in
from src.pretrained.gpen.face_model.op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d
File "/home/hou/code/e4s-main/src/pretrained/gpen/face_model/op/init.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "/home/hou/code/e4s-main/src/pretrained/gpen/face_model/op/fused_act.py", line 13, in
fused = load(
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1202, in load
return jit_compile(
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1425, in jit_compile
write_ninja_file_and_build_library(
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1537, in write_ninja_file_and_build_library
run_ninja_build(
File "/home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1824, in run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'fused': [1/2] /usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/TH -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/THC -isystem /home/hou/anaconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS -D__CUDA_NO_HALF2_OPERATORS --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/hou/code/e4s-main/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
FAILED: fused_bias_act_kernel.cuda.o
/usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/TH -isystem /home/hou/anaconda3/envs/e4s/lib/python3.8/site-packages/torch/include/THC -isystem /home/hou/anaconda3/envs/e4s/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/hou/code/e4s-main/src/pretrained/gpen/face_model/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
/usr/include/c++/11/bits/std_function.h:435:145: error: parameter packs not expanded with ‘...’:
435 | function(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:435:145: note: ‘_ArgTypes’
/usr/include/c++/11/bits/std_function.h:530:146: error: parameter packs not expanded with ‘...’:
530 | operator=(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:530:146: note: ‘_ArgTypes’
ninja: build stopped: subcommand failed.

Running environment:
gcc version 10.4.0 (Ubuntu 10.4.0-4ubuntu1~22.04)

e4s2022 / e4s Goto Github PK

e4s's People

Contributors

Stargazers

Watchers

Forkers

e4s's Issues

packages in environment at /home/user/miniconda3/envs/e4s:

Name Version Build Channel

packages in environment at C:\Users\user\miniconda3\envs\e4s:

Name Version Build Channel

Recommend Projects

Recommend Topics

Recommend Org