Blind User
Research on assistive technologies for blind users centers on improving object recognition, navigation, and data access through AI-powered systems. Current efforts focus on developing robust object recognition models that minimize errors and provide accessible feedback mechanisms, alongside creating intuitive interfaces for data contribution and control. This work is crucial for enhancing the independence and participation of blind individuals in everyday activities and for promoting inclusive data practices in AI development. Furthermore, research explores improving the robustness of these systems to various challenges, such as temporary sensor failures or adversarial attacks.
Papers
UNav: An Infrastructure-Independent Vision-Based Navigation System for People with Blindness and Low vision
Anbang Yang, Mahya Beheshti, Todd E Hudson, Rajesh Vedanthan, Wachara Riewpaiboon, Pattanasak Mongkolwat, Chen Feng, John-Ross Rizzo
Approaching English-Polish Machine Translation Quality Assessment with Neural-based Methods
Artur Nowakowski