orb_slam_2_ros
SuperPoint_SLAM
| orb_slam_2_ros | SuperPoint_SLAM | |
|---|---|---|
| 1 | 2 | |
| 587 | 504 | |
| 2.7% | - | |
| 0.0 | 1.8 | |
| 9 months ago | about 3 years ago | |
| C++ | C++ | |
| GNU General Public License v3.0 or later | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
orb_slam_2_ros
-
Odometry is changing at a faster rate with intelrealsense d435 while using orb_slam2_ros
I have compiled [this repository](https://github.com/appliedAI-Initiative/orb_slam_2_ros) on nvidia jetson nano and it has been successfully built inside my catkin workspace. I have double checked the rosnode list and topics subscribed and published by nodes and they are all good. When I launched the file `roslaunch orb_slam2_ros orb_slam2_d435_rgbd.launch` and try to see the `/orb_slam2_rgbd/pose` topic, the pose looks quite unstable and changing at a faster rate (you can see inside the video attached) while the robot is static. I have attached the tf tree in this issue [frames.pdf](https://github.com/appliedAI-Initiative/orb_slam_2_ros/files/8130863/frames.pdf) and a video link [here](https://drive.google.com/file/d/1nU13ZrD_oXfTAPqC75yUB4uY-FzfR1dC/view?usp=sharing).
SuperPoint_SLAM
-
Modular Open Source Visual SLAM
Hi everyone, I am trying to implement a VSLAM with DNN specifically the Feature Extraction module in the SLAM pipeline. Something on the lines of this repo Superpoint_SLAM , which integrates SuperPoint Feature extraction into ORB_SLAM2
-
Complete Open Source Deep Learning Implementations For V-SLAM
As you've mentioned, there are many papers on deep local feature extraction, like SuperPoint and R2D2. If you wish to use them in SLAM, you can simply replace the feature extraction module in the existing SLAM system with the deep local feature method. An example is shown here - this system uses SuperPoint as local features instead of ORB features in the original ORB-SLAM 2 pipeline. https://github.com/KinglittleQ/SuperPoint_SLAM
What are some alternatives?
Fast_RRT-ROS - RRT-ROS implementation.
rtabmap - RTAB-Map library and standalone application
Universal_Robots_ROS_Driver - Universal Robots ROS driver supporting CB3 and e-Series
openvslam - OpenVSLAM: A Versatile Visual SLAM Framework
Nox_robot - Nox robot project
open_vins - An open source platform for visual-inertial navigation research.
ov2slam - OV²SLAM is a Fully Online and Versatile Visual SLAM for Real-Time Applications
pyslam - pySLAM contains a monocular Visual Odometry (VO) pipeline in Python. It supports many modern local features based on Deep Learning.
stella_vslam - This is a unofficial fork of OpenVSLAM (https://github.com/xdspacelab/openvslam)
maplab - A Modular and Multi-Modal Mapping Framework
sloam - Semantic LIDAR odometry and mapping for cylinderical objects (e.g. trees in forests)
xivo - X Inertial-aided Visual Odometry