Section: New Results

High tech vision aid systems for low-vision patients

Using virtual reality to helping low-vision people read depending on their pathology

Participants : Marco Benzi [Université Côte d'Azur (France)] , Stéphanie Baillif [Centre hospitalier Pasteur 2 (service d’ophtalmologie, Nice, France)] , Annick Martin [”27Delvalle” (Centre d’Innovation Santé de la ville de Nice, France)] , Eric Castet [Aix-Marseille Université (CNRS, Laboratoire de Psychologie Cognitive, Marseille, France)] , Fabio Solari [University of Genoa (DIBRIS, Genoa, Italy)] , Manuela Chessa [University of Genoa (DIBRIS, Genoa, Italy)] .

By stimulating imagination, reading can be considered as the first immersive media that we are experimenting in our life. We read for leisure, to learn or to be informed. Nowadays, we read not only on printed books or newspaper but on a variety of electronic platforms (computers, tablets, phones), thus extending the possibilities to read. However, reading poses problems for almost everyone with low-vision and it is amongst the strongest need reported by patients  [55], [65]. Electronic equipments such as CCTV have offered new possibilities for the patients to tune their prefered display and many studies have been done to understand the impact of most parameters in reading performance  [50], [70], [42], [49], [65]. However, display is still highly limited by the small field of view offered by CCTVs, the navigation issues, and the fact that they are constrained to sit at their desk in order to read, thus providing a limited comfort to patients. Our goal is to investigate how virtual reality could be used to overcome these limitations and study new reading aid strategies depending on patients' pathologies.

This project received funding from Université Côte d'Azur (France), in the "Pré-maturation" call which finances actions that transform existing proof of concept into an operational laboratory prototype allowing either the realization of "robust" demonstrators or the complete experimental validation of concept (see Sec. 9.1.1).

Real-time image enhancement in virtual reality applications for low-vision people

Participants : Manuela Chessa [University of Genoa (DIBRIS, Genoa, Italy)] , Alberto Patino [University of Genoa (DIBRIS, Genoa, Italy)] , Horacio Rostro [University of Guanajuato (Guanajuato, Mexico)] , Eric Castet [Aix-Marseille Université (CNRS, Laboratoire de Psychologie Cognitive, Marseille, France)] , Fabio Solari [University of Genoa (DIBRIS, Genoa, Italy)] , Pierre Kornprobst.

In the last years, virtual reality technology has experienced a boost in affordability, and an increasing number of applications have emerged proposing new immersive 360 degrees visual content. To make this content accessible for low-vision people, one should adopt the same strategies as in traditional displays, i.e., use dedicated image enhancement methods to facilitate their interpretation. This work introduces a virtual reality application for mobile devices that implements real-time content enhancement. It is implemented as a visual search task in a set of static 360 degrees environments: the immersed user can manipulate the parameters of the enhancement algorithm in a intuitive way, using an external controller. In particular, we focus on the transform proposed by Peli et al  [69], which is based on an adaptive filter that controls the local contrast as a function of the local mean luminance of an image. Such a transform has been shown to improve recognition tasks in patients with moderate visual loss, central scotoma or cataracts. Our application is, to our knowledge, the first attempt to evaluate the impact of this image enhancement in an immersive virtual reality environment. In particular, our system allows the real time tuning of the transform, and provides all the quantitative data to analyse a posteriori users’ behaviour and how parameters may impact their performance. Designed as a game, it is perceived as more enjoyable than traditional ophthalmologic experiments. More generally, this application could be a way for low-vision people to adjust vision enhancements to their needs in everyday virtual reality applications, also for entertainment purposes

This work was presented at the Vision conference [27].

ARVIP: Augmented reality for visually impaired people

Participants : Josselin Gautier, Pierre Kornprobst, Frédéric Dosière [Bosch Visiontec (Sophia Antipolis, France)] , David Coupé [Bosch Visiontec (Sophia Antipolis, France)] .

In Biovision, we want to develop new augmented reality systems for low-vision people, to facilitate scene interpretation by enhancing important scene characteristics. Research and investigations are conducted using automotive industry HW solutions, thanks to a partnership with Bosch Visiontec (Sophia Antipolis, France, see Sec. 8.1.1).