RGBD Camera navigation

Description

We present a novel wearable RGBD camera based navigation system for the visually impaired. The system is composed of a smartphone user interface, a glass-mounted RGBD camera device, a real-time navigation algorithm, and haptic feedback system. A smartphone interface provides an effective way to communicate to the system using audio and haptic feedback. In order to extract orientational information of the blind users, the navigation algorithm performs real-time 6-DOF feature based visual odometry using a glass-mounted RGBD camera as an input device. The navigation algorithm also builds a 3D voxel map of the environment and analyzes 3D traversability. A path planner of the navigation algorithm integrates information from the egomotion estimation and mapping and generates a safe and an efficient path to a waypoint delivered to the haptic feedback system. The haptic feedback system consisting of four micro-vibration motors is designed to guide the visually impaired user along the computed path and to minimize cognitive loads. The proposed system achieves real-time performance faster than 30Hz in average on a laptop, and helps the visually impaired extends the range of their activities and improve the mobility performance in a cluttered environment. The experiment results show that navigation in indoor environments with the proposed system avoids collisions successfully and improves mobility performance of the user compared to conventional and state-of-the-art mobility aid devices.

<Fig. 1> System overview: Image/3D information acquisition device, SLAM Process, and a feedback system. 


Publication

[1] Young Hoon Lee, Gérard Medioni, Wearable RGBD indoor navigation system for the blind (Oral paper) 
     ECCV 2014 Assistive Computer Vision and Robotics Workshop, Zurich, Switzerland, Sep 2014.
[2] Young Hoon Lee, Gérard Medioni, A RGB-D camera Based Navigation for the Visually Impaired 
     RSS 2011 RGB-D: Advanced Reasoning with Depth Camera Workshop, Los Angeles, Jun 2011.


Experiment Video

[1] Wearable RGB-D navigation system: Grocery Scene

Wearable RGB-D navigation system: Grocery Scene



Awards

[1] Best Research Demo Award2nd Annual Ming Hsieh Department of Electrical Engineering Research Festival, USC
[2] FinalistCornell Cup USA, presented by Intel.

Media

[1] Reuters.

[2] USC Viterbi magazine

http://viterbi.usc.edu/news/news/2011/guide-vests-robotic.htm

 

[3] The Economic Times: How Image Processing will move the world

http://economictimes.indiatimes.com/tech/software/image-search-will-shape-the-future/articleshow/10394958.cms

 

[4] L.A. Business Journal

http://viterbi.usc.edu/assets/131/77819.pdf

 

[5] Ubergizmo: Guide vest that lets blind wearers “see” the world via tactile feedback

http://www.ubergizmo.com/2011/05/guide-vest-that-lets-blind-wearers-see-the-world-via-tactile-feedback/

 

[6] Gizmag: ‘Guide vest’ to act as eyes for the visually impaired

http://www.gizmag.com/guide-vest-visually-impaired/18722/

 

[7] MedgadgetUSC Researchers Develop “Guide Vests” for the Visually Impaired

http://medgadget.com/2011/05/usc-researchers-develop-guide-vests-for-the-visually-impaired.html

 

[8] Guide-Vest Gives Visually Impaired New Set of Eyes

http://news.discovery.com/tech/guide-vest-gives-visiually-impaired-new-set-of-eyes-110527.html

 

[9] ntdnews.com

http://therohanaurora.com/guide-vests-blind-aid/