williamljb/humanmultiview

humanmultiview's Issues

Tensor name "v_template" not found in checkpoint files

Hi,

I'm trying to run a demo. I pass with the argument --load_paths one of the files in the folder. I get:

2021-12-14 14:47:41.141066: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2021-12-14 14:47:41.141092: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2021-12-14 14:47:41.141097: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2021-12-14 14:47:41.141101: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
/Users/amirhadzic/HumanMultiView/src_ortho/../models/neutral_smpl_with_cocoplus_reg.pkl
/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/deepdish/io/hdf5io.py:428: UserWarning: This file was saved with a newer version of deepdish. Please upgrade to make sure it loads correctly.
warnings.warn('This file was saved with a newer version of '
Iteration 0
Reuse is on!
Reuse is on!
Reuse is on!
Iteration 1
Reuse is on!
Reuse is on!
Reuse is on!
Reuse is on!
Iteration 2
Reuse is on!
Reuse is on!
Reuse is on!
Reuse is on!
Restoring checkpoint /Users/amirhadzic/HumanMultiView/models/model.ckpt-55124.index..
Traceback (most recent call last):
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/runpy.py", line 174, in _run_module_as_main
"main", fname, loader, pkg_name)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/runpy.py", line 72, in _run_code
exec code in run_globals
File "/Users/amirhadzic/HumanMultiView/demo.py", line 255, in
main(config.img_paths, config.json_path)
File "/Users/amirhadzic/HumanMultiView/demo.py", line 220, in main
model = RunModel(config, 4, num_views, sess=sess)
File "src_ortho/RunModel.py", line 74, in init
self.prepare()
File "src_ortho/RunModel.py", line 169, in prepare
self.saver.restore(self.sess, self.load_path)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1560, in restore
{self.saver_def.filename_tensor_name: save_path})
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 895, in run
run_metadata_ptr)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1124, in _run
feed_dict_tensor, options, run_metadata)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1321, in _do_run
options, run_metadata)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1340, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.NotFoundError: Tensor name "v_template" not found in checkpoint files /Users/amirhadzic/HumanMultiView/models/model.ckpt-55124.index
[[Node: save/RestoreV2_282 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_arg_save/Const_0_0, save/RestoreV2_282/tensor_names, save/RestoreV2_282/shape_and_slices)]]

Caused by op u'save/RestoreV2_282', defined at:
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/runpy.py", line 174, in _run_module_as_main
"main", fname, loader, pkg_name)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/runpy.py", line 72, in _run_code
exec code in run_globals
File "/Users/amirhadzic/HumanMultiView/demo.py", line 255, in
main(config.img_paths, config.json_path)
File "/Users/amirhadzic/HumanMultiView/demo.py", line 220, in main
model = RunModel(config, 4, num_views, sess=sess)
File "src_ortho/RunModel.py", line 73, in init
self.saver = tf.train.Saver()
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1140, in init
self.build()
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1172, in build
filename=self._filename)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 688, in build
restore_sequentially, reshape)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 407, in _AddRestoreOps
tensors = self.restore_op(filename_tensor, saveable, preferred_shard)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 247, in restore_op
[spec.tensor.dtype])[0])
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/ops/gen_io_ops.py", line 663, in restore_v2
dtypes=dtypes, name=name)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/framework/op_def_library.py", line 767, in apply_op
op_def=op_def)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2630, in create_op
original_op=self._default_original_op, op_def=op_def)
File "/Users/amirhadzic/opt/anaconda3/envs/human_multi/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 1204, in init
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

NotFoundError (see above for traceback): Tensor name "v_template" not found in checkpoint files /Users/amirhadzic/HumanMultiView/models/model.ckpt-55124.index
[[Node: save/RestoreV2_282 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_arg_save/Const_0_0, save/RestoreV2_282/tensor_names, save/RestoreV2_282/shape_and_slices)]]

Which model file should I use. Simehow none of them is working..

Installation guide

Hi @williamljb thanks for such a nice and amazing work. This project is really helpful.

I'm unable to find detail about project installation, can you provide detail installation about project that where to add SMPL, pre-trained weights etc...

Thanks for your support.

Dataset for train

Hi, thanks for your nice work.
If I want to train, should I use all datasets or only the synthetic tfrecords I download ?
And should I crop the mpi_inf_3dhp dataset?

Generated mesh is not properly aligned vertically

I used four different views of an image to generate a 3d mesh. However, the mesh is tilted forward. Any advise on fixing it?

You need to specify `load_path` to load a pretrained model

Giving command
python -m demo --img_paths testImages/test.png

I was getting the above mentioned error in the title.
So I changed command to
python -m demo --img_paths testImages/test.png --load_path models/model.ckpt-55124

but now i am getting the following error :
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x80 in position 0: invalid start byte.

Any ideas?

Code getting stuck at 635 in trainer.py

I tried to run the training script, the terminal shows "Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 8694 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, compute capability: 7.5)" and gets stuck at line 644 in trainer.py(found using pdb)

Dataset Generating Code

Nice work,
I downloaded your dataset, the data generator codes as well, but there are no readme files to explain what is the generator do and how to use it.
could you please explain a litter more about what and how is your dataset generator do?

pose and shape Parameter Space Sampling

Hi，Thank you for providing the code，wonderful job.

I feel puzzled about the SMPL pose and shape param sampling mentioned in your paper section 5.1. Does uniform sampling means you not need to fit the CMU data as a mixed-gauss function, just to calculate the mean value and standard deviation to generate virtual shapes for every dims(usually as 10 dims)?

eg: numpy.random.uniform(low,high,size)

Do you provide any code or suggestion for this part? Thanks a lot.

How to integrate SMPL male model?

Hi,

Firstly, you've done a terrific work. I find this work really interesting. However, while rendering the result on the smpl mesh, I noticed that you are using neutral smpl with cocoplus regressor. Can you advice me on how it render the results using basic smpl model male?

Cloth simulation..

Hello,
Thank you for providing the code.

I downloaded your data generation code, and I want to do cloth simulation using the SMPL model. I understood that the motion data can be created by position and rotation of the pose.

However could you please explain how to use SMPL model when using arcsim?(Is it right that simply simulate motion data extracted from SMPL model to arcsim with default obstacle body mesh?)