zhengkw18 / face-vid2vid

Unofficial implementation of the paper "One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing" (CVPR 2021 Oral)
153 stars 17 forks source link

What is the version number of imageio? Thank you very much~ #8

Closed 805094591 closed 2 years ago

zhengkw18 commented 2 years ago

Any stable version of imageio for python>=3.7 should be ok

805094591 commented 2 years ago

Could you upload a requirments.yml, which include all the installed requirments and is deduced from anaconda for this project, please? Thank you very much~

zhengkw18 commented 2 years ago

Sorry, currently I do not have such an environment in my hand. I think you can choose any python>=3.7, and install the following packages without specifying the version

numpy
pillow
torch
torchvision
scikit-image
scikit-learn
imageio
pandas
805094591 commented 2 years ago

(face-vid2vid) F:\github\face-vid2vid>python evaluate.py --ckp=100 --source=out/a.jpg --driving=out Traceback (most recent call last): File "evaluate.py", line 109, in eval(args) File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\torch\autograd\grad_mode.py", line 28, in decorate_context return func(args, kwargs) File "evaluate.py", line 78, in eval video_array = [img_as_float32(io.imread(os.path.join(args.driving, frames[idx]))) for idx in range(num_frames)] File "evaluate.py", line 78, in video_array = [img_as_float32(io.imread(os.path.join(args.driving, frames[idx]))) for idx in range(num_frames)] File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\skimage\io_io.py", line 53, in imread img = call_plugin('imread', fname, plugin=plugin, plugin_args) File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\skimage\io\manage_plugins.py", line 207, in call_plugin return func(args, *kwargs) File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\skimage\io_plugins\imageio_plugin.py", line 10, in imread return np.asarray(imageio_imread(args, **kwargs)) File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\imageio\core\functions.py", line 159, in imread with imopen(uri, "ri", plugin=format) as file: File "G:\Conda_data\envs\face-vid2vid\lib\site-packages\imageio\core\imopen.py", line 277, in imopen raise err_type(err_msg) ValueError: Could not find a backend to open out\a.mp4`` with iomoderi`.

805094591 commented 2 years ago

I do not know the reason of that phenomenon~

zhengkw18 commented 2 years ago

You should notice that driving is a folder containing frames of video, not a video.

805094591 commented 2 years ago

The project runs well due to your help~ Your explanation helped me solve the problem that bothered me for a long time ~ Thank you very much~