Intelligent Image Processing
NonVisNavi: non-visual mobile navigation application for pedestrians
EuroHaptics'12 Proceedings of the 2012 international conference on Haptics: perception, devices, mobility, and communication - Volume Part II
Wearable range-vibrotactile field: design and evaluation
ICCHP'12 Proceedings of the 13th international conference on Computers Helping People with Special Needs - Volume Part II
Real-time plane segmentation and obstacle detection of 3d point clouds for indoor scenes
ECCV'12 Proceedings of the 12th international conference on Computer Vision - Volume 2
Multimodal kinect-supported interaction for visually impaired users
UAHCI'13 Proceedings of the 7th international conference on Universal Access in Human-Computer Interaction: design methods, tools, and interaction techniques for eInclusion - Volume Part I
Hi-index | 0.00 |
We present a way finding system that uses a range camera and an array of vibrotactile elements which we built into a helmet. The range camera is a Kinect 3D sensor from Microsoft that is meant to be kept stationary, and used to watch the user (i.e., to detect the person's gestures). Rather than using the camera to look at the user, we reverse the situation, by putting the Kinect range camera on a helmet for being worn by the user. In our case, the Kinect is in motion rather than stationary. Whereas stationary cameras have previously been used for gesture recognition, which the Kinect does very well, in our new modality, we take advantage of the Kinect's resilience against rapidly changing background scenery, where the background in our case is now in motion (i.e., a conventional wearable camera would be presented with a constantly changing background that is difficult to manage by mere background subtraction). The goal of our project is collision avoidance for blind or visually impaired individuals, and for workers in harsh environments such as industrial environments with significant 3-dimensional obstacles, as well as for use in low-light environments.