timobolkart / tempeh Goto Github PK

@TimoBolkart
I am getting this error, when trying to run the coarse stage inference:

loaded pretrained
resume_checkpoint(): found 1 models
Resuming progress from 600001 iteration
from model path ./runs/coarse\coarse__TEMPEH_final\checkpoints\model_00600000.pth
Traceback (most recent call last):
File "tester/test_global.py", line 78, in
Traceback (most recent call last):
File "", line 1, in
main()
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\spawn.py", line 105, in spawn_main
File "tester/test_global.py", line 75, in main
exitcode = _main(fd)
calibration_directory=calibration_directory, image_file_ext=image_file_ext, out_dir=out_dir)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\spawn.py", line 115, in _main
File "tester/test_global.py", line 53, in run
self = reduction.pickle.load(from_parent)
EOFError: Ran out of input
execute_locally(test_config_fname)
File "tester/test_global.py", line 28, in execute_locally
run(config_fname=test_config_fname)
File "C:\GANs\TEMPEH\tester\global_tester.py", line 292, in run
tester.run()
File "C:\GANs\TEMPEH\tester\global_tester.py", line 120, in run
for data in self.dataloader:
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\site-packages\torch\utils\data\dataloader.py", line 444, in iter
return self._get_iterator()
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\site-packages\torch\utils\data\dataloader.py", line 390, in _get_iterator
return _MultiProcessingDataLoaderIter(self)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\site-packages\torch\utils\data\dataloader.py", line 1077, in init
w.start()
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\process.py", line 112, in start
self._popen = self._Popen(self)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\context.py", line 223, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\context.py", line 322, in _Popen
return Popen(process_obj)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\popen_spawn_win32.py", line 89, in init
reduction.dump(process_obj, to_child)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\multiprocessing\reduction.py", line 60, in dump
ForkingPickler(file, protocol).dump(obj)
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\site-packages\torch\multiprocessing\reductions.py", line 142, in reduce_tensor
storage = tensor.storage()
File "C:\Users\RonniAdavattathuInno\anaconda3\envs\TEMPEH\lib\site-packages\torch_tensor.py", line 205, in storage
return torch._TypedStorage(wrap_storage=self._storage(), dtype=self.dtype)
NotImplementedError: Cannot access storage of SparseTensorImpl

I am trying to run tempeh in windows and the installation of the mesh viewer from MPI-IS/mesh is not getting installed on windows, and also starts giving away too many errors. Is there any replacement for the mesh viewer maybe making use of meshlab in this case? but I am new to this, don't know how to replace the "Mesh" class that you are using from psbody.mesh with meshlab? Please let me know if there is some alternative to this?

Thanks for the great work!

We have noticed variations in the orientation of different expressions among individuals in the downloaded data. This has caused challenges in processing and analyzing the data. We would like to inquire whether there is an aligned version of the data available.

I have download the test data, and unzip "data/downloads/coarse__TEMPEH_final.zip" to "runs/coarse/coarse__TEMPEH_final" , then I try to run the demo, but I can't find mesh_sampler.npz under the "./runs/coarse/coarse__TEMPEH_final/"
mesh_sampler_fname = join(self.directory_output, 'mesh_sampler.npz')
there is a file in "./runs/refinement/refinement_TEMPEH_final" but not in this path "runs/coarse/coarse__TEMPEH_final",is it the same thing, or could you provide the file for us?

I am trying to repose the sequences of the registered meshes in your dataset onto FLAME models with different shapes, for which I need facial expression parameters. However, in the downloaded "FLAME Registrations - Package 01 - 07" I can only find the .ply meshes. Where can I find the corresponding FLAME parameters?

Thank you for your impressive work!

I have my own multi-view images captured by Light Stage and have tried to use your pretrained model to generate FLAME heads, but I found that the results are completely garbled. Do I need to train the model by myself even if I want to generate heads with the same topology as yours? Or is the error due to my data not being aligned properly?

Impressive work！Can I use your pre-train model to test my own data or I need to train it myself?

Thank you for your impressive work!

I want to apply our data, multiple images (same img size as your project ) and same format calibration data(*.tka) used in your project ,to your test code with pre-trained model.
However the generated results, the generated mesh looks crash.
Is there anything I should be careful about?

Thanks

Hi,

I would be interested on having the FLAME parameters for our dataset, however, when I downloaded the "FLAME Registrations" I can only find the final reconstructions. Would it be possible to get those parameters? I am trying to repose the sequences of the registered meshes onto FLAME models with different shapes.

Thanks!!

Absolutely great work and one of my favorite papers this year! Is there any plan to further expand the dataset or make a new dataset to include dialogue with matching 3D reconstruction? Or possibly utilizing some obstruction to match in the wild dataset?