microcompunics/mobileslam

Acknowledgements:

This repository contains code from https://github.com/tum-vision/lsd_slam which we are modifying for a term project. Many thanks go out to the TUM Vision Group. The methodogy for this method, LSD-SLAM, can be found in the following paper:

LSD-SLAM: Large-Scale Direct Monocular SLAM, J. Engel, T. Schöps, D. Cremers, ECCV '14

Pre-requisites: xcode 8.0, opencv, boost, sophus, Eigen

Please make sure xcode >= 8.0 is install in your computer.

Please follow the link below to install opencv ios: http://docs.opencv.org/2.4/doc/tutorials/introduction/ios_install/ios_install.html

boost, sophus and Eigen are included in this repository.

Title:

Direct Visual Odometry on Mobile Device. Please see the demo: https://www.youtube.com/watch?v=gOveXpLiBqw.

Function:

The code has been test with iPad 2 and iPhone5. It could be running realtime 30fpsin iPad 2
and 15fps in iPhone5, which is released in 2012.

Please see the link below for detailed descriptions for each device:

iPad 2: https://www.amazon.com/Factory-Unlocked-Apple-Wi-Fi-Silver/dp/B00N086XAA/ref=sr_1_4?s=wireless&ie=UTF8&qid=1486637763&sr=1-4&keywords=Apple+iPad+2)

iPhone5: https://www.amazon.com/Apple-iPhone-16GB-Certified-Refurbished/dp/B00WZR5ULU/ref=sr_1_4?s=wireless&ie=UTF8&qid=1486637722&sr=1-4&keywords=Apple+iPhone+5.

Team:

Guanhang Wu & Jennifer Lake

Description:

We won the second best project in Advanced Computer Vision Apps class at CMU. Please see our post: Mobile_LSD_Poster.pdf and paper: Mobile_LSD_Report.pdf for implmentation and performance details.