We present in this paper the state of the art and an analysis of recent research work and achievements performed in the domain of AI-based and vision-based systems for helping blind and visually impaired people (BVIP). We start by highlighting the recent and tremendous importance that AI has acquired following the use of convolutional neural networks (CNN) and their ability to solve image classification tasks efficiently. After that, we also note that VIP have high expectations about AI-based systems as a possible way to ease the perception of their environment and to improve their everyday life. Then, we set the scope of our survey: we concentrate our investigations on the use of CNN or related methods in a vision-based system for helping BVIP. We analyze the existing surveys, and we study the current work (a selection of 30 case studies) using several dimensions such as acquired data, learned models, and human–computer interfaces. We compare the different approaches, and conclude by analyzing future trends in this domain.

We present in this article an augmented reality system called VISIT designed to deliver content about artworks to visitors. This system aims to simplify the use of augmented reality by allowing (non-computer) users to define content and place it in 3D around an artwork. VISIT is composed of two main generic elements with respect to historical or artistic content: a Web interface allows to represent in a cloud all the necessary information (sites, areas, artworks, media) to define the augmented reality content that visitors will see. Its use is simple and accessible to non-computer specialists and does not require additional software development or special hardware. A mobile augmented reality application under Android will use this content and deliver it to the visitor on his/her smartphone. The application automatically detects the presence of the artworks and then places the 3D content as specified in the Web interface. The visitors can choose the elements that interest them. The user’s actions are recorded and allow site managers to analyze their behavior. VISIT is operational and we tested it in the context of a large-scale exhibition at the Château d’Azay-Le-Rideau in France. We conclude with the progress achieved as well as the limits and perspectives of our work.

Current artificial neural networks are trained with parameters encoded as floating point numbers that occupy lots of memory space at inference time. Due to the increase in size of deep learning models, it is becoming very difficult to consider training and using artificial neural networks on edge devices such as smartphones. Binary neural networks promise to reduce the size of deep neural network models as well as increasing inference speed while decreasing energy consumption and so allow the deployment of more powerful models on edge devices. However, binary neural networks are still proven to be difficult to train using the backpropagation based gradient descent scheme. We propose to adapt to binary neural networks two training algorithms considered as promising alternatives to backpropagation but for continuous neural networks. We provide experimental comparative results for image classification including the backpropagation baseline on the MNIST, Fashion MNIST and CIFAR-10 datasets in both continuous and binary settings. The results demonstrate that binary neural networks can not only be trained using alternative algorithms to backpropagation but can also be shown to lead better performance and a higher tolerance to the presence or absence of batch normalization layers.

The SCULPTURE3D research project is a collaborative and multidisciplinary laboratory experiment financed by the Centre-Val de Loire region between 2014 and 2018, which aims to explore what 3D brings to the study of sculpture and to interpret the region’s heritage of sculptures from the Renaissance Era. It is an association between researchers from the University of Tours, working with the Centre d’études supérieures de la Renaissance (CESR) and with the Fundamental and Applied Computer Science of Tours (LIFAT), and various heritage institutions (DRAC Centre, Museum of Fine Arts of Tours, CMN, Château de Blois…) as well as a team of experts on innovative softwares (CETU ILIAD3), and the new small business Kulturae’com, two innovative entities who got their start through that project. The thirty digital objects created with the help of a 3D optical scanner is the first key part of a virtual museum of 3D sculptures from the Renaissance Era in the Val de Loire, intended both for researchers and the general public. Similarly, the Sculpturoscope, la Vierge à l’Enfant du réel au virtuel meta-exposition (at the Museum of Fine Arts of Tours, from May 25^th to September 10^th 2018) made the results of scientific research available to the general public by using digital medias as both a tool of investigation and mediation.