Semantic labelling to aid navigation in prosthetic vision

Assessing people with visual impairments’ access to information, awareness and satisfaction with high-tech assistive technology

Assistive technology (AT) devices are designed to help people with visual impairments (PVIs) perform activities that would otherwise be difficult or impossible. Devices specifically designed to assist PVIs by attempting to restore sight or substitute it for another sense have a very low uptake rate. This study, conducted in England, aimed to investigate why this is the case by assessing accessibility to knowledge, awareness, and satisfaction with AT in general and with sensory restoration and substitution devices in particular. From a sample of 25 PVIs, ranging from 21 to 68 years old, results showed that participants knew where to find AT information; however, health care providers were not the main source of this information. Participants reported good awareness of different ATs, and of technologies they would not use, but reported poor awareness of specific sensory substitution and restoration devices. Only three participants reported using AT, each with different devices and varying levels of satisfaction. The results from this study suggest a possible breakdown in communication between health care providers and PVIs, and dissociation between reported AT awareness and reported access to AT information. Moreover, awareness of sensory restoration and substitution devices is poor, which may explain the limited use of such technology.

Unifying Terrain Awareness for the Visually Impaired through Real-Time Semantic Segmentation

Navigational assistance aims to help visually-impaired people to ambulate the environment safely and independently. This topic becomes challenging as it requires detecting a wide variety of scenes to provide higher level assistive awareness. Vision-based technologies with monocular detectors or depth sensors have sprung up within several years of research. These separate approaches have achieved remarkable results with relatively low processing time and have improved the mobility of impaired people to a large extent. However, running all detectors jointly increases the latency and burdens the computational resources. In this paper, we put forward seizing pixel-wise semantic segmentation to cover navigation-related perception needs in a unified way. This is critical not only for the terrain awareness regarding traversable areas, sidewalks, stairs and water hazards, but also for the avoidance of short-range obstacles, fast-approaching pedestrians and vehicles. The core of our unification proposal is a deep architecture, aimed at attaining efficient semantic understanding. We have integrated the approach in a wearable navigation system by incorporating robust depth segmentation. A comprehensive set of experiments prove the qualified accuracy over state-of-the-art methods while maintaining real-time speed. We also present a closed-loop field test involving real visually-impaired users, demonstrating the effectivity and versatility of the assistive framework.