classner / generating_people Goto Github PK
View Code? Open in Web Editor NEWOfficial code repository for the paper "A Generative Model for People in Clothing".
Home Page: http://files.is.tuebingen.mpg.de/classner/gp/
License: Other
Official code repository for the paper "A Generative Model for People in Clothing".
Home Page: http://files.is.tuebingen.mpg.de/classner/gp/
License: Other
Hello :)
Could you provide your code, which should be located in
~/up/pose/pose_deepercut.py
I can't find this code. Thanks.
how do I generate image based on clothing and person as an input ?
Hi @classner ,
I check closed issue about smpl dataset link, but it looks like it happened again.
http://smpl.is.tue.mpg.de/signup
I can't sign up this smpl site so I can't download smpl dataset and library.
When I click it sign up button, it says " We're sorry, but something went wrong. ".
Could you help to show me how to download these stuffs?
Thanks,
JH Moon.
Thank you for answering my question last time and I have another question.
To set up the data, i run the scripts iteratively .
But when run 05_run_fits.sh, i got this error.
IndexError: index 4 is out of bounds for axis 1 with size 3
Also, you told me that 'pose_deepercut.py' code is in 'https://github.com/classner/up'
but there is no 'pose_deepercut.py' and only other code, so i run the 'pose.py'.
Thanks.
Hi, are there any documentation about how to use the generate_conditioned script? Specifically I'd like to specify clothing colour as described in the paper. Thanks
I ran the code with
./generate 20
I got following error
2018-02-22 23:53:03,842 __main__ INFO Running on host: aclab-H97M-D3H
2018-02-22 23:53:03,843 __main__ INFO Running mode `sample` for experiment `experiments/config/LSM`.
Traceback (most recent call last):
File "./run.py", line 354, in <module>
cli()
File "/home/jun/miniconda3/envs/py2.7/lib/python2.7/site-packages/click/core.py", line 722, in __call__
return self.main(*args, **kwargs)
File "/home/jun/miniconda3/envs/py2.7/lib/python2.7/site-packages/click/core.py", line 697, in main
rv = self.invoke(ctx)
File "/home/jun/miniconda3/envs/py2.7/lib/python2.7/site-packages/click/core.py", line 895, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/home/jun/miniconda3/envs/py2.7/lib/python2.7/site-packages/click/core.py", line 535, in invoke
return callback(*args, **kwargs)
File "./run.py", line 78, in cli
path.join(exp_name, 'config.py')) # source this one error?
File "experiments/config/LSM/config.py", line 1
../template/config.py
^
SyntaxError: invalid syntax
It seems the syntax can't be recognized
../template/config.py
I am running it on
Ubuntu 16.04
python 2.7 (conda)
tensorflow 1.4
Thanks.
When I run the code "generation/tools/04_run_deepercut.sh", I find that I have no the module "~/git/up" stated in the "config.py" at Line 13. How can I get the exact code?
I find the project DeepCut(http://pose.mpi-inf.mpg.de/#code) as described in the paper. But it doesn't provide the code used in this project.
Hi @classner ,
When I load the model "basicModel_neutral_lbs_10_207_0_v1.0.0.pkl" and 'regressors_locked_normalized_hybrid.npz' in ~/git/up/models/3D directory, I notice that the dimension of betas is 10 + 1 =11, while _REGRESSOR is input with the dimension of 301x20. This leads to mismatch the dimension in the following codes in capsule_man.py line 132 - 136:
pad = ch.concatenate((wrt_betas, ch.zeros(n_shape_dofs-len(wrt_betas)), ch.ones(1)))
lengths = pad.dot(length_regs)
rads = pad.dot(rad_regs)
Could you explain what is the meaning of 10 and 301x20?
Best,
Hai
I am getting this error:
Exception: /home/drive/generate_people/generating_people-master/generation/data/people/full contains no files for this mode
I saw you've already answered this question .But I'm still confused where to download the tf-record data. Could you please answer my question ?
I am getting this error:
Exception: /home/drive/generate_people/generating_people-master/generation/data/people/full
contains no files for this mode
Which are the files that should be put in this folder and how to get it?
Hi @classner ,
Thanks for you interesting work. I try to download the dataset and smpl, but the link is broken and nothing is responsed.
smpl: http://smpl.is.tue.mpg.de/downloads
dataset: http://files.is.tuebingen.mpg.de/classner/gp/
I also try to run scripts in generation/tools, but where do you download the data and their segmentation? I can not generate data by scripts without segmentation folder.
Could you help to show me how to download these stuffs?
Thanks,
Hai
Hi @classner,
Thank you for updating the code. I re-run ./05_run_fits.sh, but there is another error of out of index in the following code in SphereCollision.py:
def get_objective(self):
print('get objective')
return ch.sum((ch.exp(-((ch.sum((self.sph_v[self.ids0]-
self.sph_v[self.ids1])**2,
axis=1))/(self.radiuss))/2.))2).5
self.sph_v only has 3 elements, while self.ids0 and self.ids1 has larger value. Could you help to see the image below?
Best,
Hai
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.