Computer vision technologies to assist blind persons

CCNY Data Science, Networks, and Biology Seminar

Time and place

3:30 PM on Wednesday, November 9th, 2016; NAC 7/219

Yingli Tuan (CCNY Electrical Engineering)

Abstract

In this talk, I will introduce our research of applying computer vision technologies to assist visually impaired or blind persons. In particular, this talk will focus on scene text detection extraction and indoor navigation. Text information widely exists in natural scene and serves as a significant indicator for many vision-based applications. Automatic extracting text information from natural scene images is still a challenging task due to complex background outliers and variations of text patterns. We have developed a prototype navigation system in Android platform by combining computer vision and robotics SLAM technologies to assist blind users in navigating across multiple floors.