A wearable face recognition system for individuals with visual impairments

Abstract

This paper describes the iCare Interaction Assistant, an assistive device for helping the individuals who are visually impaired during social interactions. The research presented here addresses the problems encountered in implementing real-time face recognition algorithms on a wearable device. Face recognition is the initial step towards building a comprehensive social interaction assistant that will identify and interpret facial expressions, emotions and gestures. Experiments conducted for selecting a face recognition algorithm that works despite changes in facial pose and illumination angle are reported. Performance details of the face recognition algorithms tested on the device are presented along with the overall performance of the system. The specifics of the hardware components used in the wearable device are mentioned and the block diagram of the wearable system is explained in detail.

Low-cost face biometry for visually impaired users

“…Unlike [29] we aim at a simple procedure to gather the training data, ideally something that the user could do by himself, possibly with the help of a friend. Our model may start from a very small set of images of the person (say 10), that could be easily captured, and no requirement on the person pose or the scene illumination is made.…”

“…An improvement, both from the point of view of efficiency and performance, would be to apply a feature selection phase to reduce the size of the initial data description. This approach has been previously explored by the authors for the case of face validation and showed a superior performance with respect to other well known dimensionality reduction methods, such as PCA (we remind the reader that PCA was adopted by [29] as the most appropriate method within a similar application to the one we are discussing). Experiments reported in [3], that we briefly report here for completeness, shown a superiority of local methods with respect to PCA: while PCA achieved an acceptance rate (AR) of 0.68 and a rejection rate (RR) of 0.7, a local method based on LBP allowed us to achieve AA = 0.87 and RR = 0.94.…”

“…Among recent projects it is worth mentioning the iCare project 3 developing a set of digital tools for blind people, and in particular proposing automatic face recognition to address prosopagnosia also known as face blindness. In [29] a face recognition method to improve socialization skills of visually impaired population is proposed. To the best of our knowledge [29] is the only project addressing the same issue we deal with in this paper.…”

“…In [29] a face recognition method to improve socialization skills of visually impaired population is proposed. To the best of our knowledge [29] is the only project addressing the same issue we deal with in this paper.…”

An Iranian Cash Recognition Assistance System For Visually Impaireds

“…Kurzweil reader [9], iCare [10] (uses a wearable camera for imaging and a CPU chip for computation) and some other devices have been developed too, each has their advantages and disadvantages, but the common property of them is that the user most carry a device everywhere. Some of them are bulky and expensive too.…”