/KPConv_for_DALES

Primary LanguagePythonMIT LicenseMIT

Intro figure

Update 27/04/2020: PyTorch implementation available. With SemanticKitti, and Windows supported.

This repository contains the implementation of Kernel Point Convolution (KPConv), a point convolution operator presented in our ICCV2019 paper (arXiv).

Update 03/05/2019, bug found with TF 1.13 and CUDA 10. We found an internal bug inside tf.matmul operation. It returns absurd values like 1e12, leading to the apparition of NaNs in our network. We advise to use the code with CUDA 9.0 and TF 1.12. More info in issue #15

Installation

A step-by-step installation guide for Ubuntu 16.04 is provided in INSTALL.md. Windows is currently not supported as the code uses tensorflow custom operations.

TO USE IN DALES DATASET

Use the convert.py to convert the DALES ascii ply file to bin ply file. copy the convert.py to the location of the ascii ply files and run it.
Utilize requirements and conda_env For conda env creation you can use :

conda create --name <name of your choice> --file envlist.txt

and for pip

pip install -r pipenvlist.txt

MY RESULTS IN DALES DATASET

The result of the testing is uploaded here and the test_accuracy.py was used to calculate F1 scores which gave this result with Final Avg micro F1 : 0.97 | Avg macro F1: 0.72. This result was far better than any other point convolutions I had used.