Run demo.npy, click "predict", enter the weight and height, and click OK to flash back and show that the process has ended. The exit code is - 1073740791 (0xc0000409).

I get this error.

Using TensorFlow backend.
2020-01-31 11:57:05.217716: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory
2020-01-31 11:57:05.217852: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory
2020-01-31 11:57:05.217885: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Traceback (most recent call last):
  File "train.py", line 8, in <module>
    from fancyimpute import MICE
ImportError: cannot import name 'MICE'

is it possible that the model outputs (theta & beta) parameters ( pose & shape ) in a (.npz & .npy) file format instead of .ply file ?
if not . can you please guide us ?
thank you.

hi, I was able to bring up your app but it's frozen. I couldn't manipulate with any UIs at all.

Here is the steps I set up the environment.

I created a virtual 3.5.5 environment on anaconda and install the following packages using pip.

Ubuntu 16.04
numpy 1.14.5
scipy 1.1.0
fancyimpute 0.3.2
mayavi 4.6.1
openpyxl 2.5.5
PyOpenGL 3.1.0
pyqt 4.11.4
vtk 8.1.1

There seems to be one conflict in your requirements. As stated on the homepage of Mayavi just supports PyQt5 in Python3, but the requirements have both Python3, Mayavi and PyQt4.

Could you provide a bit more detail about the version of the packages (Mayavi and PyQt) you installed?

Thanks

hi @zengyh1900 @rockbot

user must enter some its body shape parameters in order to get the 3D its body model?
if I have the mash values of specific person how can I extract its body measurement ?

If the input is a 3D obj file, Could you give the measurments of the human body? thanks

Can you manually change the vertices on the 3d model by using javascript or python? Does the text dataset automate and suggest height, weight, and BMI proportions? I'm just curious. I'd like it if the user doesn't like the dataset formality then the user can alter the vertices on the 3d model to control the mass material (not the weight).

Hey! Thanks for your work! Can you tell me what P2M means in the code? How to generate it?

Do you have support for VTK version 9 and up?

I have a model of vertex count = 6890 and triangle face count = 13776, i am trying to load this data but i figured out i have to manually set body control points. Please also tell me about template.txt file? what is it about and how can I manually set control points? much appreciated.

Hi, first of all i want to express my gratitude for this great project that u have implemented and given this for our reference.

I am having some trouble in compiling the code

this is the error i am receiving can u pls check this out and let me know what i need to do to rectify this error

Hi! If i want to use another model which has different toplogy, how can i get the body control points?

Hello,
could you tell me the method to measure the body_control_points.txt
or the means of the each clumn information?

Thank you.

blue_measurement_guide.pdf

This is not an actual issue, just posting here so other people may find this useful. The measurements described in the paper used the above guide as a reference, so you may now correctly relate the names of the measurements with how they look on the 3D human model.

Hi, great works!
After playing with the demo. I want to train several SPRING models for a better understanding. I also realized that most of the *.npy files have to be manually added into the obj2npy function which will call several pre-defined functions.

However, for instance, if I have 10 SPRING models in the obj/female folder. Which one should I employ to compute the normals? Say the "vertex" in obj2npy is of shape (len(file_list), V_NUM, 3), but the one required by the compute_normals(vertex, facet) appears to be of shape (V_NUM, 3).

Here's the code for the revised obj2npy:

def obj2npy(label="female"):
print(' [**] begin obj2npy about %s... '%label)
start = time.time()
OBJ_DIR = os.path.join(DATA_DIR, "obj")
obj_file_dir = os.path.join(OBJ_DIR, label)
file_list = os.listdir(obj_file_dir)

# load original data
vertex = []
for i, obj in enumerate(file_list):
    sys.stdout.write('\r>> Converting %s body %d'%(label, i))
    sys.stdout.flush()
    f = open(os.path.join(obj_file_dir, obj), 'r')
    j = 0
    for line in f:
        if line[0] == '#':
            continue
        elif "v " in line:
            line.replace('\n', ' ')
            tmp = list(map(float, line[1:].split()))
            vertex.append(tmp)
            j += 1
        else:
            break

facet = convert_template()

vertex = np.array(vertex).reshape(len(file_list), V_NUM, 3)

normals = compute_normals(vertex[0], facet)
np.save(open(os.path.join(DATA_DIR, "normals.npy"), "wb"), normals)
print("finish compute normals")



for i in range(len(file_list)):
    v_mean = np.mean(vertex[i,:,:], axis=0)
    vertex[i,:,:] -= v_mean
mean_vertex = np.array(vertex.mean(axis=0)).reshape(V_NUM, 3)
std_vertex = np.array(vertex.std(axis=0)).reshape(V_NUM, 3)

facet = np.load(open(os.path.join(DATA_DIR, "facet.npy"),"rb"))
# loading deform-based data
[d_inv_mean, deform, mean_deform, std_deform] = load_d_data(vertex, facet, label)
# calculating deform-based presentation(PCA)
[d_basis, d_coeff, d_pca_mean, d_pca_std] = get_d_basis(deform, label)#female_d_coeff.npy

np.save(open(os.path.join(DATA_DIR, "%s_vertex.npy"%label), "wb"), vertex)
np.save(open(os.path.join(DATA_DIR, "%s_mean_vertex.npy"%label), "wb"), mean_vertex)
np.save(open(os.path.join(DATA_DIR, "%s_std_vertex.npy"%label), "wb"), std_vertex)

#Adding extra stuff ...
cp = convert_cp()
[measure, mean_measure, std_measure, t_measure] = convert_measure(cp, vertex, facet, label) #生成%s_measure.npy

# calculate global mapping from measure to deformation PCA coeff
get_m2d(d_coeff, t_measure, label)  # "%s_m2d.npy"

# cosntruct the related matrix A to change deformation into vertex
get_d2v_matrix(d_inv_mean, facet, label)# %s_d2v.npz

#mask = np.load(open(os.path.join(DATA_DIR, "mask.npy"), "rb"))
# get color dict & mask
[PART, mask] = get_map(facet)
# local map matrix: measure->deform
local_matrix(mask, deform, measure, label)#"%s_local.npy"


print('\n [**] finish obj2npy in %fs.' %(time.time() - start))
return [vertex, mean_vertex, std_vertex, file_list]

However, the normals.npy generated in this way will cause an error in reshaper.py, line 182
return [x, -self.normals, self.facets - 1]
TypeError: bad operand type for unary -: 'map'

If I use your original normals.npy , I can pass this and generate 100 new models with code in line 217. However, even if I am using your normals.npy, and replaced the rest of the *.npy files with my training result, the demo.py will not deliver the correct result.

As you can see, I am getting a meat ball for setting the female body as weight == 65 kg and height == 166cm.

Any idea how to fix this? Tons of thanks in advance~~~!!!

Hello,

I am wondering if your code could be used on a Linux machine?
Because I tried to install pyqt4 and didn't manage to do it unfortunately!

hi, I m trying to use your code but it always occurs the following error.


It seems unable to create this file cp.npy. Is there something wrong with my directory?
looking forward to your reply.

how to install ctypes on ubuntu 16.04

Hello,
Thanks for your great repo. I have a question, is it possible for 3D humans to learn a mapping of transformed points in 3D human to 2D image space?

Thank you

hi,
I was able to bring up your app but it's frozen. I couldn't manipulate with any UIs at all.
i also have same issue. i am using linux. i am able to get GUI , but it is frozen in 4 to 5 seconds and unable adjust the corresponding values on GUI.

input weight 58 ,height 170 , chick predict and then click save the result , weight is 31 and height is 169

@1900zyh Would it be possible to add additional parameters such as calf, ankle, and forearm?

The model seems to do well but I would like to fine tune some of those parameters

While running the body_utils.py It needs some numpy files

File "body_utils.py", line 535, in train
vertex = np.load(open(os.path.join(DATA_DIR, "%s_vertex.npy"%label),"rb"))
IOError: [Errno 2] No such file or directory: '/home/administrator/3D-Human-Body-Shape/data/male_vertex.npy'
but, in data folder dosn't have those files.

Hi, you assigned colors to your body_part.obj. How you obtained it. Can we retrieve corresponding vertices and faces related to particular body part???
Thank you

Hello,

I have some difficulties to install this python project with the requirement.txt file which is given.
Using Python3.5.6 of Anaconda, some dependencies cannot be solved.
And using the last version of Python (but removing the specific version in the requirement.txt file and modifying FancyImpute(MICE) as described by AlyanQ), I succeed to install the project. But when I run the training, I am facing an error in rfe_multiprocess.

Can someone describe a procedure to have a code running without issue using Anaconda ?

Regards,

Nabla

If you install and try to run this repo in 2021, you will get errors for FancyImpute and the MICE imputer module. FancyImputer's MICE imputer has been deprecated and been merged into sci-kit learns Iterative Imputer and some of the syntax relating to the module has changed as well. This is the current fix to get it to work correctly.

In line https://github.com/1900zyh/3D-Human-Body-Shape/blob/1c71ea7f19fb87a0881d51415b4344a904ed01eb/src/reshaper.py#L4

#from fancyimpute import MICE
from fancyimpute import IterativeImputer as MICE

In line https://github.com/1900zyh/3D-Human-Body-Shape/blob/1c71ea7f19fb87a0881d51415b4344a904ed01eb/src/reshaper.py#L120

#tmp = solver.complete(tmp)
tmp = solver.fit_transform(tmp)

Hope this is useful to someone.

I am confused with max 30 size. how to measure this from inches

run on ubuntu16.04 with python3.5, and install requirements.txt
python demo.py
Using TensorFlow backend.
Traceback (most recent call last):
File "demo.py", line 7, in
from maya_widget import *
File "/home/julian/workd/git_baseline/3D-Human-Body-Shape/src/maya_widget.py", line 8, in
from PyQt4 import QtGui, QtCore
ValueError: PyCapsule_GetPointer called with incorrect name

Inside the reshaper file function MICE is not defined.
Kindly recheck it once...

Hi,
Great work!

I tried running the GUI using demo.py but it gave me the following error,
mlab.triangular_mesh(v[:, 0], v[:, 1], v[:, 2], f) TypeError: string indices must be integer

So, I tried to run it manually using reshaper.py,
that gives an error message of female_measure.npy not found. It is not found anywhere in the repo too.

Can you please clarify on this and also can you give a sample of the OBJ that is generated using this repo?

EMAIL: [email protected]

Is there a web version I could try?

mlab.triangular_mesh(v[:, 0], v[:, 1], v[:, 2], f)
TypeError: string indices must be integers, not tuple
Traceback (most recent call last):
File "demo.py", line 228, in
show_app()
File "demo.py", line 222, in show_app
win = HumanShapeAnalysisDemo()
File "demo.py", line 23, in init
self.viewer3D = MayaviQWidget(container)
File "/home/administrator/3D-Human-Body-Shape/src/maya_widget.py", line 72, in init
layout.addWidget(self.ui)
TypeError: QBoxLayout.addWidget(QWidget, int stretch=0, Qt.Alignment alignment=0): argument 1 has unexpected type 'PySide.QtGui.QWidget'
QClipboard: Unable to receive an event from the clipboard manager in a reasonable time
[acerE15:19042] *** Process received signal ***
[acerE15:19042] Signal: Segmentation fault (11)
[acerE15:19042] Signal code: Invalid permissions (2)
[acerE15:19042] Failing at address: 0x5ce183b
[acerE15:19042] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x11390)[0x7f7d6ea7d390]
[acerE15:19042] [ 1] [0x5ce183b]
[acerE15:19042] *** End of error message ***
Segmentation fault (core dumped)

is there any one that help me to configure this git it create value Error that indicate reshaping issue when i run train .py file

Hi

Running this in a Colab Notebook.

I get :
/content/3D-Human-Body-Shape/src
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-root'
qt.qpa.screen: QXcbConnection: Could not connect to display
Could not connect to any X display.

Can't I test it in Colab Notebook?

Hello there,

Thanks for open-sourcing this research work.
Can I get some more clarifications for all the 19 measurements that you had given as inputs?
what does natural waist, maximum hip, natural waist rise mean and gluteal hip?

Thank you.

Hi, In 3D-Human-Body-Shape, inappropriate dependency versioning constraints can cause risks.

Below are the dependencies and version constraints that the project is using

cython==0.29.7
numpy==1.16.1
scipy==1.2.1
openpyxl==2.6.2
ecos==2.0.7.post1
vtk==8.1.2
traits==5.1.1
cvxpy==1.0.21
fancyimpute==0.3.2
mayavi==4.7.1
pyopengl==3.1.0
pyqt5==5.10.1
scikit-learn==0.21.3

The version constraint == will introduce the risk of dependency conflicts because the scope of dependencies is too strict.
The version constraint No Upper Bound and * will introduce the risk of the missing API Error because the latest version of the dependencies may remove some APIs.

After further analysis, in this project,
The version constraint of dependency numpy can be changed to >=1.8.0,<=1.23.0rc3.
The version constraint of dependency scipy can be changed to >=0.12.0,<=1.7.3.
The version constraint of dependency openpyxl can be changed to >=1.5.7,<=1.8.6.
The version constraint of dependency scikit-learn can be changed to >=0.9,<=0.21.3.

The above modification suggestions can reduce the dependency conflicts as much as possible,
and introduce the latest version as much as possible without calling Error in the projects.

The invocation of the current project includes all the following methods.

numpy.linalg.det
numpy.linalg.svd
numpy.linalg.inv
numpy.linalg.norm

scipy.sparse.coo_matrix
scipy.sparse.coo_matrix.transpose

openpyxl.Workbook.save
openpyxl.Workbook
openpyxl.Workbook.get_active_sheet

sklearn.feature_selection.RFE
sklearn.linear_model.LinearRegression
sklearn.feature_selection.RFE.fit

self.radio2.setFont
PyQt5.QtWidgets.QAction.__init__
traitsui.api.Item
numpy.zeros.repeat
sys.stdout.write
results.np.array.reshape
self.reset_button.setFont
self.setCentralWidget
self.set_slider
data.np.array.reshape.append
int
maya_widget.IndexedQSlider.setRange
self.setWindowTitle
fancyimpute.MICE.complete
self.mainBox.addLayout
M.transpose.dot
numpy.zeros
scipy.sparse.linalg.spsolve
tmplist.append
line.strip
self.valueChanged.connect
self.body.mapping
i.self.rfemat.dot
utils.M_NUM.body_num.results.np.array.reshape.transpose
self.pre_button.setToolTip
self.statusBar
PyQt5.QtWidgets.QLineEdit
self.label.append
convert_cp.append
PyQt5.QtWidgets.QHBoxLayout
self.radio1.setChecked
PyQt5.QtWidgets.QLabel.setFixedHeight
utils.V_BASIS_NUM.v_basis.np.array.reshape
vertex.mean.np.array.reshape
HumanShapeAnalysisDemo.show
event.accept
tmp.index.tmp.index.tmp.replace
line.replace
numpy.load.std
self.pre_dialog.show
numpy.savez
len
self.t_measure.copy
self.editList.text
i.self.editList.text
numpy.std
line.strip.split
self.radio1.toggled.connect
dets.np.array.reshape.append
numpy.sum
traits.api.on_trait_change
PyQt5.QtWidgets.QSpinBox
PyQt5.QtWidgets.QMainWindow.__init__
v3.v2.v1.assemble_face.dot
PyQt5.QtWidgets.QAction
get_d_basis
L_tosave.append
tmp.np.array.reshape
openpyxl.Workbook.save
measure_list.append
maya_widget.MayaviQWidget
self.valueChangeForwarded.emit
t_measure.transpose
build_equation
utils.calc_measure
self.pre_button.clicked.connect
train
spinBox.valueChanged.connect
DATA_DIR.os.path.join.open.read
Reshaper.mapping
i.self.slider.valueChangeForwarded.disconnect
measure.std.np.array.reshape
self.vertices.astype
self.radio2.toggled.connect
PyQt5.QtWidgets.QLabel.setFont
numpy.linalg.det
openpyxl.Workbook.get_active_sheet
normals.append
PyQt5.QtWidgets.QSlider.__init__
mat.sum
self.d_synthesize
measure_list.np.array.reshape
numpy.linalg.svd
reshaper.Reshaper.mapping
isinstance
d2v.d2v.transpose.dot.tocsc
Reshaper
self.pre_dialog.close
multiprocessing.Pool.map
open.close
numpy.zeros.mean
idx.t_measure.reshape
self.menuBar.addMenu
numpy.zeros.transpose
self.pre_dialog.setLayout
numpy.row_stack
PyQt5.QtWidgets.QPushButton.setFont
numpy.mean
self.set_radio
PyQt5.QtWidgets.QRadioButton
self.radio1.isChecked
data.np.array.reshape
open
construct_coeff_mat
self.mapping_mask
tmp.replace.split
PyQt5.QtWidgets.QSpinBox.setRange
self.visualization.edit_traits
PyQt5.QtWidgets.QWidget.show
PyQt5.QtWidgets.QApplication
i.self.local_mat.dot
flag.sum
obj2npy.std
tmp.max
self.button_box.addWidget
self.viewer3D.setSizePolicy
self.local_mat.append
convert_measure.std
mat.tmp.np.row_stack.transpose
v1.v0.np.cross.dot
obj2npy
openpyxl.Workbook
scipy.sparse.coo_matrix
multiprocessing.Pool.join
utils.V_BASIS_NUM.v_basis.np.array.reshape.transpose
numpy.cross
rfe_local
self.myact.emit
self.m2d.dot
numpy.matmul
maya_widget.IndexedQSlider
i.self.editList.clear
numpy.dot.copy
self.viewer3D.predict
round
map
PyQt5.QtWidgets.QWidget.setLayout
i.face.vertexNormalLists.append
float
i.self.editList.setText
self.setToolTip
PyQt5.QtWidgets.QWidget.setWindowTitle
self.body.get_predict
time.time
print
utils.save_obj
sklearn.feature_selection.RFE
utils.PART.index
rfe.support_.np.array.reshape
multiprocessing.Pool
vertex.np.array.reshape
slider.valueChanged.connect
t_mask.repeat.repeat
in_data.copy
i.self.rfemask.np.array.reshape
maya_widget.myAction
line.split
PyQt5.QtWidgets.QLabel.setText
list
self.reset_button.clicked.connect
get_inv_mean
i.self.slider.valueChangeForwarded.connect
build_equation.transpose
results.ele.ele.np.array.reshape
set.index
PyQt5.QtWidgets.QGridLayout
self.update
dialogOK.clicked.connect
calc_measure
mayavi.mlab.clf
convert_measure.mean
self.ui.setParent
maya_widget.IndexedQSlider.setStatusTip
numpy.linalg.norm
numpy.zeros.std
load_d_data
convert_measure
self.d2v.transpose
self.slider.append
show_app
numpy.sqrt
traitsui.api.View
self.minimum
set.replace
numpy.load.mean
maya_widget.IndexedQSlider.setFixedWidth
d2v.transpose.dot
mayavi.mlab.triangular_mesh
self.mainBox.addWidget
numpy.column_stack
weight.np.array.reshape
self.dialogBox.addLayout
obj2npy.append
PyQt5.QtWidgets.QWidget
PyQt5.QtWidgets.QAction.setStatusTip
utils.M_NUM.utils.F_NUM.results.ele.ele.np.array.reshape.transpose
menubar.addMenu.addAction
PyQt5.QtWidgets.QApplication.exec_
PyQt5.QtGui.QFont
utils.D_BASIS_NUM.d_basis.np.array.reshape.transpose
format
self.spin.append
self.editList.setText
self.lu.solve
abs
reshaper.Reshaper
PyQt5.QtWidgets.QPushButton
self.set_menu
multiprocessing.Pool.close
convert_cp
deform.flat.np.array.reshape
mode.myact.connect
self.triggered.connect
self.visualization.update_plot
convert_template
wb.get_active_sheet.cell
set
self.set_button
self.mapping_global
run
PyQt5.QtWidgets.QVBoxLayout.addWidget
self.pre_dialog.setWindowTitle
float.tolist
sys.stdout.flush
self.set_dialog
dets.np.array.reshape
self.reset_button.setStatusTip
PyQt5.QtWidgets.QDialog
mayavi.core.ui.api.SceneEditor
PyQt5.QtWidgets.QVBoxLayout.setContentsMargins
os.path.join
color_list.append
self.maximum
output.tmp.np.column_stack.transpose
self.menuBar
numpy.float64
measure.mean.np.array.reshape
save.triggered.connect
self.d2v.self.d2v.transpose.dot.tocsc
obj2npy.transpose
PyQt5.QtCore.pyqtSignal
traits.api.Instance
numpy.array
numpy.linalg.inv
PyQt5.QtWidgets.QHBoxLayout.addWidget
self.viewer3D.select_mode
scipy.sparse.linalg.splu
utils.D_BASIS_NUM.d_basis.np.array.reshape
PyQt5.QtWidgets.QLabel.setFixedWidth
os.listdir
obj2npy.mean
enumerate
self.d2v.transpose.dot
self.pre_button.setFont
numpy.load
slider.valueChangeForwarded.connect
numpy.save
PyQt5.QtWidgets.QWidget.__init__
PyQt5.QtWidgets.QVBoxLayout
Visualization
i.self.mask.np.array.reshape
PyQt5.QtWidgets.QAction.setShortcut
HumanShapeAnalysisDemo
range
multiprocessing.Pool.starmap
PyQt5.QtWidgets.QLabel
random.sample
self.resize
tmp.min
open.write
sys.exit
self.editList.append
PyQt5.QtWidgets.QVBoxLayout.setSpacing
self.box.addLayout
vertex.std.np.array.reshape
i.self.slider.setValue
PyQt5.QtWidgets.QGridLayout.addWidget
sklearn.feature_selection.RFE.fit
numpy.dot
clearButton.clicked.connect
fancyimpute.MICE
dets.np.array.reshape.ravel
self.radio1.setFont
set.add
self.imputate
exit.triggered.connect
d.np.array.reshape
assemble_face
scipy.sparse.coo_matrix.transpose
self.statusBar.showMessage
sklearn.linear_model.LinearRegression
self.mapping_rfemat

@developer
Could please help me check this issue?
May I pull a request to fix it?
Thank you very much.

hello,
after python train.py it will take too much time. how much take time to train a model? . stuck in

loading male deformation of facet 24999

Using TensorFlow backend.
2020-01-31 14:27:51.964071: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory
2020-01-31 14:27:51.964203: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory
2020-01-31 14:27:51.964219: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

Hi,
I am running the code successfully and i understand the concept.
I want to segments the r obj obtained.i.e, giving colors to each, for face, hands, legs, body,etc., how to do that. and also how to obtain particular vertices and faces of a specific body part.
By knowing this i can do furthor process. Thank you...