DPRNet : Deep 3D Point based Residual Network for Semantic Segmentation and Classification of 3D Point Clouds

https://ieeexplore.ieee.org/abstract/document/8721650

Usage

This code is tested in Ubuntu 16.04 LTS with CUDA 8.0 and Tensorflow-gpu==1.4. First of all you need to compile convolutional operators as follow:

cd tf_ops/conv3p/

chmod 777 tf_conv3p_compile.sh

./tf_conv3p_compile.sh -a

if you are using tensorflow-gpu==1.4 or above then you might want to try compiling with tf_conv3p_compile_tf14.sh instead. It fixes some include paths due to nsync_cv.h, and set the flag _GLIBCXX_USE_CXX11_ABI=0 to make it compatible to libraries compiled with GCC version earlier than 5.1.

After successfully compiling the convolution operators you can start DPRNet training as follow:

To train object classification, execute

python train_modelnet40.py [epoch]

To evaluate, execute

python eval_modelnet40.py [epoch]

Similar procedure is required for scene segmentation task. By default 'epoch' is 0. You can resume the training by passing epoch number in the above command.

Training Data

Data folder contains links of the datasets for both classification and semantic segmentation task.

Dependencies

This code includes the following third party libraries and data:

• Scaled exponential linear units (SeLU) for self-normalization in neural network.

• ModelNet40 data from PointNet

• Some other utility code from Pointwise CNN

• h5py

Acknowledgemets

The code for convolution operators, Training and evaluation borrowed from [Pointwise CNN]