GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → jhultman → continuous-fusion

jhultman / continuous-fusion

Licence: MIT license
(ROS) Sensor fusion algorithm for camera+lidar.

Programming Languages

C++
36643 projects - #6 most used programming language

Labels

rosperceptionlidarsensor-fusionpointcloudkitticontinuous-fusionlidar-camera

Projects that are alternatives of or similar to continuous-fusion

lopocs
Migrated to: https://gitlab.com/Oslandia/lopocs
Stars: ✭ 78 (+200%)
Mutual labels:  lidar, pointcloud
Loam noted
loam code noted in Chinese（loam中文注解版）
Stars: ✭ 455 (+1650%)
Mutual labels:  lidar, pointcloud
awesome-lidar
😎 Awesome LIDAR list. The list includes LIDAR manufacturers, datasets, point cloud-processing algorithms, point cloud frameworks and simulators.
Stars: ✭ 217 (+734.62%)
Mutual labels:  lidar, pointcloud
CarND-Extended-Kalman-Filter-P6
Self Driving Car Project 6 - Sensor Fusion(Extended Kalman Filter)
Stars: ✭ 24 (-7.69%)
Mutual labels:  lidar, sensor-fusion
the-Cooper-Mapper
An open source autonomous driving research platform for Active SLAM & Multisensor Data Fusion
Stars: ✭ 38 (+46.15%)
Mutual labels:  perception, sensor-fusion
fusion-ekf
An extended Kalman Filter implementation in C++ for fusing lidar and radar sensor measurements.
Stars: ✭ 113 (+334.62%)
Mutual labels:  lidar, sensor-fusion
lt-mapper
A Modular Framework for LiDAR-based Lifelong Mapping
Stars: ✭ 301 (+1057.69%)
Mutual labels:  lidar, pointcloud
Tools Merge Image PointCloud
Project the PointCloud to the image & Generate the LiDAR PointCloud with color.
Stars: ✭ 39 (+50%)
Mutual labels:  pointcloud, kitti
Pyicp Slam
Full-python LiDAR SLAM using ICP and Scan Context
Stars: ✭ 155 (+496.15%)
Mutual labels:  lidar, pointcloud
Loam velodyne
Laser Odometry and Mapping (Loam) is a realtime method for state estimation and mapping using a 3D lidar.
Stars: ✭ 1,135 (+4265.38%)
Mutual labels:  lidar, pointcloud
PointPainting
This repository is an open-source PointPainting package which is easy to understand, deploy and run!
Stars: ✭ 152 (+484.62%)
Mutual labels:  lidar, pointcloud
annotate
Create 3D labelled bounding boxes in RViz
Stars: ✭ 104 (+300%)
Mutual labels:  lidar, pointcloud
learning-topology-synthetic-data
Tensorflow implementation of Learning Topology from Synthetic Data for Unsupervised Depth Completion (RAL 2021 & ICRA 2021)
Stars: ✭ 22 (-15.38%)
Mutual labels:  sensor-fusion, kitti
sensor-fusion
Filters: KF, EKF, UKF || Process Models: CV, CTRV || Measurement Models: Radar, Lidar
Stars: ✭ 96 (+269.23%)
Mutual labels:  lidar, sensor-fusion
SpinNet
[CVPR 2021] SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration
Stars: ✭ 181 (+596.15%)
Mutual labels:  pointcloud, kitti
Sc Lego Loam
LiDAR SLAM: Scan Context + LeGO-LOAM
Stars: ✭ 332 (+1176.92%)
Mutual labels:  lidar, pointcloud
Las Rs
Read and write ASPRS las files, Rust edition.
Stars: ✭ 27 (+3.85%)
Mutual labels:  lidar, pointcloud
OpenMaterial
3D model exchange format with physical material properties for virtual development, test and validation of automated driving.
Stars: ✭ 23 (-11.54%)
Mutual labels:  perception, lidar
lodToolkit
level-of-details toolkit(LTK). Convert osgb lod tree to 3mx tree. Convert pointcloud in ply/las/laz/xyz to 3mx/osgb tree.
Stars: ✭ 81 (+211.54%)
Mutual labels:  lidar, pointcloud
WS3D
Official version of 'Weakly Supervised 3D object detection from Lidar Point Cloud'(ECCV2020)
Stars: ✭ 104 (+300%)
Mutual labels:  lidar
View All Similar Projects ➔

Continuous fusion

C++ implementation (in ROS) of continuous fusion algorithm from Deep Continuous Fusion for Multi-Sensor 3D Object Detection. I can share a fast, vectorized SciPy version if there is any interest (update: see issues page).

Fusion

Comments on coordinate systems:

Points x_velo in velodyne coords are sent to points x_image
in the image plane of camera 2 using (7) in Geiger et al.:

    x_image = P * R * T * x_velo,

where:

    x_velo  - point in velodyne coords (x, y, z, 1).
    T       - velodyne coords to unrectified cam0 coords.
    R       - unrectified cam0 coords to rectified cam 0 coords,
    P       - rectified cam0 coords to image plane of cam2,
    x_image - point in image of cam2 (u, v, 1).

(We use left-multiplying active/alibi coordinate transformations.)

Diagram borrowed from paper (we skip MLP):

Fusion

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].