Skip to content

ModelNet10 shape classification implemented with tensorflow

Notifications You must be signed in to change notification settings

Sooram/ModelNet-3dCNN-tensorflow

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Dataset: ModelNet10

http://modelnet.cs.princeton.edu/#

Data preparation

https://github.com/guoguo12/modelnet-cnn3d_bn

.off -(1.voxelize)-> .binvox -(2.read)-> Numpy array

  1. Voxelize
    Read in [.off] file and change it into [.binvox] file of binary voxel data , using binvox program.

  2. Read
    Read [.binvox] file into Numpy array with (30,30,30) shaped elements, using binvox_rw.py and prepare_data.py.

Network architecture

very simple: input-ccp-ccp-dense-output Overview

Visualization

[.binvox] file can be viewed here.

Study(paper review)

https://docs.google.com/presentation/d/183N5PUKAzACj1IkOT4K3F5m1iN-x-KaPIufhYXkqw5E/edit?usp=sharing

About

ModelNet10 shape classification implemented with tensorflow

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages