gradslam

Open source

gradslam is an open source differentiable dense SLAM library for PyTorch

1.4K stars0 likes1 views

About

gradslam is an open source differentiable dense SLAM library for PyTorch

Overview

gradslam is a fully differentiable dense SLAM framework. It provides a repository of differentiable building blocks for a dense SLAM system, such as differentiable nonlinear least squares solvers, differentiable ICP (iterative closest point) techniques, differentiable raycasting modules, and differentiable mapping/fusion blocks. One can use these blocks to construct SLAM systems that allow gradients to flow all the way from the outputs of the system (map, trajectory) to the inputs (raw color/depth images, parameters, calibration, etc.).

python

rgbdimages = RGBDImages(colors, depths, intrinsics)
slam = PointFusion()
pointclouds, recovered_poses = slam(rgbdimages)
pointclouds.plotly(0).show()

Installation

Requirements

pytorch>=1.6.0 (for other pytorch versions see here)

Using pip (Experimental)

pip install gradslam

Install from GitHub

pip install 'git+https://github.com/gradslam/gradslam.git'

Install from local clone (Recommended)

bash

git clone https://github.com/krrish94/chamferdist.git
cd chamferdist
pip install .
cd ..
git clone https://github.com/gradslam/gradslam.git
cd gradslam
pip install -e .[dev]

Verifying the installation

To verify if gradslam has successfully been built, fire up the python interpreter, and import!

python

import gradslam as gs
print(gs.__version__)

You should see the version number displayed.

Citing gradslam

If you find gradslam useful in your work, and are writing up a report/paper about us, we'd appreciate if you cited us. Please use the following bibtex entry.

@inproceedings{gradslam,
  title={gradSLAM: Dense SLAM meets automatic differentiation},
  author={{Krishna Murthy}, Jatavallabhula and Saryazdi, Soroush and Iyer, Ganesh and Paull, Liam},
  booktitle={arXiv},
  year={2020},
}