Thank you very much for your work. Now I can convert the mesh in the shapenet data to a watertight mesh through ManifoldPlus, but I don’t know how to set up the data set structure for training. If you can give some advice, I would appreciate it.

Sir, I've created ./bounce/bounce0.txt and ./bounce/face.npy under {path_to_rp_dennis_posed_004_OBJ}. Now, I try to train, but get RunTimeError:

(tokurEnv) hamit@hamit-MS-7B49:~/Softwares/environments/PIFu$ python -m apps.train_shape --dataroot /home/hamit/Softwares/environments/PIFu/tempImages --random_flip --random_scale --random_trans
/home/hamit/Softwares/environments/PIFu/lib/data/TrainDataset.py:102: UserWarning: loadtxt: Empty input file: "/home/hamit/Softwares/environments/PIFu/tempImages/val.txt"
  var_subjects = np.loadtxt(os.path.join(self.root, 'val.txt'), dtype=str)
train data size:  180
test data size:  360
initialize network with normal
Using Network:  hgpifu
Traceback (most recent call last):
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 761, in _try_get_data
    data = self._data_queue.get(timeout=timeout)
  File "/usr/lib/python3.6/multiprocessing/queues.py", line 104, in get
    if not self._poll(timeout):
  File "/usr/lib/python3.6/multiprocessing/connection.py", line 257, in poll
    return self._poll(timeout)
  File "/usr/lib/python3.6/multiprocessing/connection.py", line 414, in _poll
    r = wait([self], timeout)
  File "/usr/lib/python3.6/multiprocessing/connection.py", line 911, in wait
    ready = selector.select(timeout)
  File "/usr/lib/python3.6/selectors.py", line 376, in select
    fd_event_list = self._poll.poll(timeout)
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/_utils/signal_handling.py", line 66, in handler
    _error_if_any_worker_fails()
RuntimeError: DataLoader worker (pid 12928) is killed by signal: Killed. 

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/usr/lib/python3.6/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/hamit/Softwares/environments/PIFu/apps/train_shape.py", line 183, in <module>
    train(opt)
  File "/home/hamit/Softwares/environments/PIFu/apps/train_shape.py", line 90, in train
    for train_idx, train_data in enumerate(train_data_loader):
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 345, in __next__
    data = self._next_data()
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 841, in _next_data
    idx, data = self._get_data()
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 808, in _get_data
    success, data = self._try_get_data()
  File "/home/hamit/Softwares/environments/tokurEnv/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 774, in _try_get_data
    raise RuntimeError('DataLoader worker (pid(s) {}) exited unexpectedly'.format(pids_str))
RuntimeError: DataLoader worker (pid(s) 12928) exited unexpectedly

PyTorch version : 1.4.0

Hi. I am having some trouble with my understanding. I hope you can enlighten and will truly appreciate it! You use spatial sampling with the ground truth mesh. Does this not mean you have a ground truth 3d occupancy field that is incomplete? For corresponding pixels without a groundtruth inside/outside prediction, will their feature vectors and z-values still be fed through PIFu?

I noticed in an earlier paragraph that you mentioned the use of bilinear sampling to obtain the feature vectors. How is the purpose of this sampling different from spatial sampling used with the ground truth meshes?

I also checked out the script you attached in the previous issue. surface_points, _ = trimesh.sample.sample_surface(mesh, 4 * self.num_sample_inout) sample_points = surface_points + np.random.normal(scale=self.opt.sigma, size=surface_points.shape) Is this above spatial sampling? I do not see a direct connection to the spatial sampling described in the paper so I just want to confirm.

Thank you for you patience. I have picked this paper to try and learn as much as possible. Hopefully I'm not annoying you. I look forward to your reply.

Using your source code on geometry we generated, we got this error during the training "Error cannot marching cubes". This error appears when the iteration is finished, it restarts the iteration again. Is this expected or something is wrong with our settings?

Thanks again for sharing your source.

/content/PIFu/lib/colab_util.py in <module>()
     16 
     17 # Data structures and functions for rendering
---> 18 from pytorch3d.structures import Meshes, Textures
     19 from pytorch3d.renderer import (
     20     look_at_view_transform,

ImportError: cannot import name 'Textures'

Thanks for your excellent work. When I using your colab demo, it gives me the above error. Could you please tell me how to fix the error? Sorry for the dumb question. I searched for it but nothing helps.

Dear Sir; For single view Pifu training, I almost got the data format: [image_tensor -> our images( 4-dim ) sample_tensor -> for mesh points, calib_tensor , labels (1 for inside points zero for the outsides)]

But I didn't find any solution for multi-view format. Should I create an image_tensor parameter for each view? #31 here, u mentioned about transform matrixes. When we creating data with "apps.render_data" we just rotate the object, right? Should I find the rotation matrices between the images and give them to the function? Can you give more information about data format when you are available? Sincerely

Thank you for your great work. I trained the model with my own data. When I eval with only shape net, there is no problem. But when I eval with both net_G and net_C, errors occurred: RuntimeError: Error(s) in loading state_dict for ResBlkPIFuNet: Missing key(s) in state_dict: "image_filter.model.2.weight", "image_filter.model.2.bias", "image_filter.model.5.weight", "image_filter.model.5.bias", "image_filter.model.8.weight", "image_filter.model.8.bias", "image_filter.model.10.conv_block.2.weight", "image_filter.model.10.conv_block.2.bias", "image_filter.model.10.conv_block.6.weight", "image_filter.model.10.conv_block.6.bias", "image_filter.model.11.conv_block.2.weight", "image_filter.model.11.conv_block.2.bias", "image_filter.model.11.conv_block.6.weight", "image_filter.model.11.conv_block.6.bias", "image_filter.model.12.conv_block.2.weight", "image_filter.model.12.conv_block.2.bias", "image_filter.model.12.conv_block.6.weight", "image_filter.model.12.conv_block.6.bias", "image_filter.model.13.conv_block.2.weight", "image_filter.model.13.conv_block.2.bias", "image_filter.model.13.conv_block.6.weight", "image_filter.model.13.conv_block.6.bias", "image_filter.model.14.conv_block.2.weight", "image_filter.model.14.conv_block.2.bias", "image_filter.model.14.conv_block.6.weight", "image_filter.model.14.conv_block.6.bias", "image_filter.model.15.conv_block.2.weight", "image_filter.model.15.conv_block.2.bias". Unexpected key(s) in state_dict: "image_filter.model.1.bias", "image_filter.model.4.bias", "image_filter.model.7.bias", "image_filter.model.10.conv_block.1.bias", "image_filter.model.10.conv_block.5.bias", "image_filter.model.11.conv_block.1.bias", "image_filter.model.11.conv_block.5.bias", "image_filter.model.12.conv_block.1.bias", "image_filter.model.12.conv_block.5.bias", "image_filter.model.13.conv_block.1.bias", "image_filter.model.13.conv_block.5.bias", "image_filter.model.14.conv_block.1.bias", "image_filter.model.14.conv_block.5.bias", "image_filter.model.15.conv_block.1.bias", "image_filter.model.15.conv_block.5.bias".

Hello! @shunsukesaito awesome work.

My question is the following: I have several images of the same person from different angles (the images were created at the same time), how can I interpolate one common object of better quality from these images? So that this object takes into account all projections. Is it possible to implement this feature using this solution?

I would be grateful if you can help me with this. Thanks!

Hi, it would be great if you could document how to use this method in the multi-view setting and also provide the model needed. :)

Thank you, amazing work.

Hi,

I have my models scaled to unit box in meshlab. I have trained this data with sigma=5 in which i observed no change in the error. I have trained the same data with sigma 0.1 where I observed error decreasing. However, marching cubes error still persists even after 15th epoch. May I know if any changes to code should be incorporated from changing meshes to unit box?

Thanks,

Hello shunsuke, I trained a netG using my own data, but the reconstructed meshes are not as good as the result of your released model. I'm not sure whether it is the data problem or did I miss something. Can you show some details like how is the MSE error going during your training process? Thank you so much!

Bumps pillow from 9.0.0 to 9.3.0.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Hello.

We are collecting data to train new models. I'm asking you because I want to know how much data i need.

The paper said that RenderPeople used about 500 scaned data for training, but I wonder why it is 500.

Is there a reason why it should be 500?

Hi, in the paper you report multi-view results, do you train the multi-view PIFu with opt.random_multiview=True? I find that if this flag is set to be true, then the model cannot generate satisfactory results, while good results can be acheived by training with fixed angle distance.

my environment is a Linux server machine, which os is ubuntu18.04, when I run this command: python -m apps.train_color --dataroot Traindata --num_sample_inout 0 --num_sample_color 5000 --sigma 0.1 --random_flip --random_scale --random_trans an error occurred, like this: ValueError: Sample larger than population or is negative

Can you help me solve this problem?

So,I tried training PiFU on the sample dennis obj,after training for 70 epochs this is my current result.Is this what you were getting for that particular object. Compared to the actual dennis object,I feel that a lot of the features especially in the face,hand and front view section were missed.Plus the entire model looks sort of blocky.

Dennis reconstruction my result The ground truth dennis model is as follows Note in the dennis folder I have only included one image which is rp_dennis_posed_004_A

Hi, Thanks for open sourcing the work.I am looking into the method and would like to apply it to a custom dataset. Based on the csv files in PiFUHD and the colab example,I am assuming that this should be the following format of the data

Train
Model_person_1
.obj file
model_person_1.jpg
Model person_1_pose_1
.obj file
model_person_1_pose_1.jpg
Model_person_2
contains an obj file and jog file

TEST
Same format as train

Second question,Let's say I have already created the model using Blender and I would like to use PiFU to just estimate the texture of the clothes,as such can I achieve this by executing train_color_model.py If yes,for evaluation what steps should be followed,because in the current demo script supports both model construction and transfer color.