dimatura / binvox-rw-py Goto Github PK

View Code? Open in Web Editor NEW

232.0 232.0 101.0 76 KB

Python module to read and write .binvox files.

License: GNU General Public License v3.0

Python 100.00%

binvox-rw-py's People

Contributors

Stargazers

Watchers

Forkers

shapecompletion3d joel-simon weijing martijnpieters kristofe crlab wkentaro thunguyenphuoc zeitgeistqian belledon archive-git-repo sy1990 daerduocarey carlos-ea raahii dvicini lepy science-code pclausen sunnycd ryanglambert qixuxiang lucabergamini moocxingsh pyqqc itzikmalkiel mrdavidyu lmescheder ahmedalkordy peteflorence tangaggie andreanicastro ketulsh zaiqiangwu nextmap bkakilli edwinyi nnu-gisa jiangyanjavawin xiaodongww bpkcl630 sigmaadrich humaolin ainurbegalinova qiushoujie harris2001 shoujieqiu docvaughan magicknight diggle1 jingnanshi rmichnovicz attaullah-brohi hengle yinyunie kunalmgupta zivzone attashe mbapepp hippogriff zzzeus chesterzhang yefanzhou fwilliams littlefishe chenhsing charlesfry guangyaoshi razekmaiden huiwen99 qiuhongannawei devriesmatt jing-lun iven-wu learnerlyh jiunyen-ching yucheng21 clemacort noncomposmentis areyesan arasbacho antoturi enesvs brotherta faridyagubbayli okkann rhys-newbury rahatzamancse duducosmos

binvox-rw-py's Issues

AttributeError: 'str' object has no attribute 'write'

Hi,
I modified some codes in 'write' of binvox_rw.py to resolve the error - 'AttributeError: 'str' object has no attribute 'write' , when executing the command - model.write('dilated.binvox').
Hope it helps.

def write(voxel_model, fp):
""" Write binary binvox format.

Note that when saving a model in sparse (coordinate) format, it is first
converted to dense format.

Doesn't check if the model is 'sane'.

"""
if voxel_model.data.ndim==2:
    # TODO avoid conversion to dense
    dense_voxel_data = sparse_to_dense(voxel_model.data, voxel_model.dims)
else:
    dense_voxel_data = voxel_model.data
print('fp : ', fp)
with open(fp, 'w') as fw:
    fw.write('#binvox 1\n')
    fw.write('dim '+' '.join(map(str, voxel_model.dims))+'\n')
    fw.write('translate '+' '.join(map(str, voxel_model.translate))+'\n')
    fw.write('scale '+str(voxel_model.scale)+'\n')
    fw.write('data\n')
    if not voxel_model.axis_order in ('xzy', 'xyz'):
       raise ValueError('Unsupported voxel model axis order')

    if voxel_model.axis_order=='xzy':
       voxels_flat = dense_voxel_data.flatten()
    elif voxel_model.axis_order=='xyz':
       voxels_flat = np.transpose(dense_voxel_data, (0, 2, 1)).flatten()

# keep a sort of state machine for writing run length encoding
    state = voxels_flat[0]
    ctr = 0
    for c in voxels_flat:
       if c==state:
           ctr += 1
        # if ctr hits max, dump
           if ctr==255:
              fw.write(chr(state))
              fw.write(chr(ctr))
              ctr = 0
       else:
          # if switch state, dump
           fw.write(chr(state))
           fw.write(chr(ctr))
           state = c
           ctr = 1
# flush out remainders
       if ctr > 0:
         fw.write(chr(state))
         fw.write(chr(ctr))

how do we set parameters for bounding box?

please update with new version of numpy

Hello thanks for the software! Can you please update to work with new Numpy to handle their bool data types.
Otherwise you get an error for using np.bool

The write function doesn't work with Python 3

If you open a file in binary 'wb' mode and call the write function you get some errors:

fp = open("test.binvox", "wb")
voxels.write(fp)

To fix the error, I have:

called ".encode('ascii')" on all the header string lines
called ".tobytes()" on the numpy bool state variable: state.tobytes()
called ".to_bytes(1, byteorder='little')" on the ctr variable: ctr.to_bytes(1, byteorder='little')

Not sure if it is the best fix, but at least it worked for me.

2 failures when executing binvox

Hi.
I still have this two failures when executing the binvox file. Anybody has an idea or knows the solution for that ?
I would appreciate it a lot. Thanks !

**********************************************************************
File ".../binvox.py", line 29, in __main__
Failed example:
    m1.scale
Expected:
    41.133000000000003
Got:
    41.133
**********************************************************************
File ".../binvox.py", line 60, in __main__
Failed example:
    np.all(ms.data[:, np.lexsort(ms.data)] == data_ds[:, np.lexsort(data_ds)])
Expected:
    True
Got:
    False
**********************************************************************
1 items had failures:
   2 of  18 in __main__
***Test Failed*** 2 failures.

I was unable to convert from the .off to .binvox format.

Hi, I'm new and I apologize for my lack of experience. I have a problem how to convert .off file to .binvox format

Directory: C:....\ModelNet10\bathtub\test

-a----        14.08.2014     07:50          73975 bathtub_0107.off
-a----        14.08.2014     07:50          14938 bathtub_0108.off
-a----        14.08.2014     07:51         428850 bathtub_0109.off
-a----        14.08.2014     07:51         338935 bathtub_0110.off
-a----        14.08.2014     07:51         470698 bathtub_0111.off
-a----        14.08.2014     07:50         117702 bathtub_0112.off
-a----        14.08.2014     07:50          48393 bathtub_0113.off
-a----        14.08.2014     07:51         143730 bathtub_0114.off
-a----        14.08.2014     07:51         191575 bathtub_0115.off
-a----        14.08.2014     07:51          15594 bathtub_0116.off
-a----        14.08.2014     07:51         311254 bathtub_0117.off
-a----        14.08.2014     07:50         286763 bathtub_0118.off
-a----        14.08.2014     07:50         304718 bathtub_0119.off
-a----        14.08.2014     07:50          42378 bathtub_0120.off
-a----        14.08.2014     07:51         118111 bathtub_0121.off
-a----        14.08.2014     07:50         411882 bathtub_0122.off
-a----        14.08.2014     07:51          56186 bathtub_0123.off
-a----        14.08.2014     07:50         157223 bathtub_0124.off
-a----        14.08.2014     07:50         383230 bathtub_0125.off
-a----        14.08.2014     07:50           9986 bathtub_0126.off
-a----        14.08.2014     07:50          36448 bathtub_0127.off
-a----        14.08.2014     07:51         216594 bathtub_0128.off
-a----        14.08.2014     07:51          30617 bathtub_0129.off
-a----        14.08.2014     07:51         112736 bathtub_0130.off
-a----        14.08.2014     07:51          56037 bathtub_0131.off
-a----        14.08.2014     07:50          33398 bathtub_0132.off
-a----        14.08.2014     07:50         154185 bathtub_0133.off
-a----        14.08.2014     07:50          54843 bathtub_0134.off
-a----        14.08.2014     07:50         605524 bathtub_0135.off
-a----        14.08.2014     07:50         155996 bathtub_0136.off
-a----        14.08.2014     07:50           8926 bathtub_0137.off
-a----        14.08.2014     07:50         138728 bathtub_0138.off
-a----        14.08.2014     07:50          73863 bathtub_0139.off
-a----        14.08.2014     07:50          31482 bathtub_0140.off
-a----        14.08.2014     07:51        2774386 bathtub_0141.off
-a----        14.08.2014     07:50         157153 bathtub_0142.off
-a----        14.08.2014     07:51          50353 bathtub_0143.off
-a----        14.08.2014     07:50          25148 bathtub_0144.off
-a----        14.08.2014     07:50         304505 bathtub_0145.off
-a----        14.08.2014     07:51         575635 bathtub_0146.off
-a----        14.08.2014     07:50          68845 bathtub_0147.off
-a----        14.08.2014     07:50          34615 bathtub_0148.off
-a----        14.08.2014     07:52        5238305 bathtub_0149.off
-a----        14.08.2014     07:51        4085485 bathtub_0150.off
-a----        14.08.2014     07:50         243233 bathtub_0151.off
-a----        14.08.2014     07:51          39501 bathtub_0152.off
-a----        14.08.2014     07:51         310543 bathtub_0153.off
-a----        14.08.2014     07:50          75717 bathtub_0154.off
-a----        14.08.2014     07:51         390411 bathtub_0155.off
-a----        14.08.2014     07:51         619530 bathtub_0156.off

binvox -c -d 24 -t binvox bathtub_0107.off

output

--- [binvox] mesh voxelizer, version 1.35, build #794 on 2021/10/26 11:37:14, running on Windows
--- written by Patrick Min, 2004-2021

  using carving method only
Windows does not allow widths below 116, increasing dimension to 192 and performing -down as needed
  Note: setting -dmin to 1
loading model file...
MeshFileIdentifier::*create_mesh_file(bathtub_0107.off)
OffMeshFile::load
  starting to read 1568 vertices
  1500 lines read
  read 1568 vertices
  3391 lines read
Read 1820 faces, 1568 vertices.
Mesh memory use is 18142 KB (17 MB)
  Mesh::normalize, bounding box: [-22.8962, -46.0039, -19.47, 1] - [22.8962, 46.0039, 19.47, 1]
    longest length: 92.0078
  normalization transform:
  (1) translate [22.8962, 46.0039, 19.47, 1], (2) scale 0.0108686, (3) translate [0, 0, 0]
  got mesh bounding box [-22.8962, -46.0039, -19.47, 1] - [22.8962, 46.0039, 19.47, 1]

voxel model dimension: 192
GLwindow::init_lighting
Graphics card and driver information:
Vendor:   Intel
Renderer: Intel(R) UHD Graphics 630
Version:  4.6.0 - Build 30.0.101.1153
Voxels::init(192, 192, 192, 0)
Voxelizer constructor
SimpleMeshView constructor
starting voxelization...
Voxelizer::carve_voxelize
  doing x direction
  doing y direction
  doing z direction
Voxelizer::fill_voxels
  Voxels::clear
Voxels::downsample from 7077888 to 884736
Voxels::downsample from 884736 to 110592

I didn't see bathtub_0107.binvox format

Converting .step file to binvox

I am trying to convert .step file to binvox. Is it possible to convert? Can anybody give suggestions?

Thank you

binvox write does not work for python3

Nice little tool; saved me some work but this bug took some time to figure out...

The write-method does not work for python3. The problem is handling of writing bytes in Python3 has changed.

I am not sure if I was doing something wrong but I was trying to push a fix but it did not work. I am very new to git so if you can help me push the change I would appreciate it. It is probably something simple I have forgotten. What i did:

git clone https://github.com/dimatura/binvox-rw-py.git
# did my changes
git add binvox_rw.py
git commit -m "Issues with writting with Python3. Fixed by using import struct; struct.pack('B',byte_value)"
git push origin HEAD:master
remote: Permission to dimatura/binvox-rw-py.git denied to pclausen.
fatal: unable to access 'https://github.com/dimatura/binvox-rw-py.git/': The requested URL returned error: 403

Thanks. Below the patch.

diff --git a/binvox_rw.py b/binvox_rw.py
index 704fc1f..6151a1e 100644
--- a/binvox_rw.py
+++ b/binvox_rw.py
@@ -1,3 +1,4 @@
+# -*- coding: utf-8 -*-
 #  Copyright (C) 2012 Daniel Maturana
 #  This file is part of binvox-rw-py.
 #
@@ -62,6 +63,7 @@ True
 """
 
 import numpy as np
+import struct
 
 class Voxels(object):
     """ Holds a binvox model.
@@ -228,6 +230,16 @@ def sparse_to_dense(voxel_data, dims, dtype=np.bool):
     #TODO ensure this is right when dims are not all same
     #"""
     #return x*(dims[1]*dims[2]) + z*dims[1] + y
+    
+def bwrite(fp,s):
+    fp.write(s.encode())
+    
+
+def write_pair(fp,state, ctr):
+    fp.write(struct.pack('B',state))
+    fp.write(struct.pack('B',ctr))
+    
+    
 
 def write(voxel_model, fp):
     """ Write binary binvox format.
@@ -244,11 +256,11 @@ def write(voxel_model, fp):
     else:
         dense_voxel_data = voxel_model.data
 
-    fp.write('#binvox 1\n')
-    fp.write('dim '+' '.join(map(str, voxel_model.dims))+'\n')
-    fp.write('translate '+' '.join(map(str, voxel_model.translate))+'\n')
-    fp.write('scale '+str(voxel_model.scale)+'\n')
-    fp.write('data\n')
+    bwrite(fp,'#binvox 1\n')
+    bwrite(fp,'dim '+' '.join(map(str, voxel_model.dims))+'\n')
+    bwrite(fp,'translate '+' '.join(map(str, voxel_model.translate))+'\n')
+    bwrite(fp,'scale '+str(voxel_model.scale)+'\n')
+    bwrite(fp,'data\n')
     if not voxel_model.axis_order in ('xzy', 'xyz'):
         raise ValueError('Unsupported voxel model axis order')
 
@@ -265,19 +277,16 @@ def write(voxel_model, fp):
             ctr += 1
             # if ctr hits max, dump
             if ctr==255:
-                fp.write(chr(state))
-                fp.write(chr(ctr))
+                write_pair(fp, state, ctr)
                 ctr = 0
         else:
             # if switch state, dump
-            fp.write(chr(state))
-            fp.write(chr(ctr))
+            write_pair(fp, state, ctr)
             state = c
             ctr = 1
     # flush out remainders
     if ctr > 0:
-        fp.write(chr(state))
-        fp.write(chr(ctr))
+        write_pair(fp, state, ctr)
 
 if __name__ == '__main__':
     import doctest

What the x,y,z coordinates locations denotes (mid point of voxel or anything else)?

What are these x,y,z coordinates locations, outputting by binvox_rw.read_as_coord_array(f) function. Are they mid point of any voxel or any vertices of a particular voxel. I want to find all the 8 vertices of a voxel. So please let me know the conversion formulas.

Bug in the write function when state change after 255 identical states

I guess the write function is bugged if you have exactly 255 zero to write, and then the next state the 256th character is a 1 (or vice versa, i.e. if you switch state after exactly 255 count).

The fix is easy: in the else of the "if c==state", add another test before dumping, the test should be: "if ctr > 0:", otherwise you will write a state with zero count, and actually may shift the whole data.

The whole code becomes:

for c in voxels_flat:
        if c==state:
            ctr += 1
            # if ctr hits max, dump
            if ctr==255:
                fp.write(chr(state))
                fp.write(chr(ctr))
                ctr = 0
        else:
            # if switch state, dump
            if ctr > 0:
                fp.write(chr(state))
                fp.write(chr(ctr))
            state = c
            ctr = 1

Can you add one more property function to resize the resolution for voxels?

The default binvox resolution for shapenet is 128128128 for example, and for some use we may need a different size (like 323232). Can you add that resize function please? It would be really helpful

Retreiving size of each voxel

Hello,

I have just created a binvox file and managed to load it using your program :)

I had one small question: How can I retrieve the size of each voxel ? I need to know the volume of each cube so knowing the edge length would be enough.

Thanks for your help,

cannot import

I am working on google colab and trying to run an already existing code. However, the import of binvox_rw is not working and even with the manual download option, there is an error with the download and it does not work. I could not find any solution yet. Can you help me, please?