Hiring Institution
Institut Mines Telecom Atlantique (IMT)

PhD-Awarding Institutions
Institut Mines Telecom Atlantique (IMT)
The University of Adelaide (UoA)

Download and read the full position description before applying

Option 1

From lab experiments to on-field applications: the effect of movement and motion on perception in Augmented Reality

Augmented Reality (AR) allows to present digital images, objects and environments superimposed on the real world. Though, due to technical limitations, the rendering of these digital elements remains different from the real elements.
Current research in the field focuses on making these digital elements as close as possible to the real ones and to measure the differences in users’ perception caused by these differences. Those controlled studies are often conducted in labs, with carefully designed environments and procedures, to ensure their accuracy and reproducibility. However, these strict conditions lead the users to perform tasks that are often very different from real AR applications. In particular, users are usually seated, with restricted movement capabilities. However, movement and motion, whether self-movement including proprioception and external motion, are known to play an important role in perception.
This project aims to investigate differences that can be observed between lab experiments and actual AR application, in order to provide a way to transpose AR research regarding basic perception issue from labs to ecological applications. The focus of this work will be to understand how motion cues from the real environment may conflict with those from the AR environment, and if so how to resolve these conflicts. Applications for this work could improve the provision of information to people with low vision, for example, while moving in the environment or while driving.

Option 2

From lab experiments to on-field applications: the effect of light and shadows on perception in Augmented Reality

Augmented Reality (AR) allows to present digital images, objects and environments superimposed on the real world. Though, due to technical limitations, the rendering of these digital elements remains different from the real elements. Transparency, occlusion, luminance, or shadow defect can appear and are still a challenge to completely overcome.
Current research in the field focuses on making these digital elements as close as possible to the real ones and to measure the differences in users’ perception caused by these differences. Those controlled studies are often conducted in labs, with carefully designed environments and procedures, to ensure their accuracy and reproducibility. However, in real applications, the real background onto which are displayed the augmented elements can highly vary. Differences in luminance, shadows, brightness or even colours can occur, making the final rendering and results highly dependent on this environment. Furthermore, when interacting with AR elements in a real environment may require a suite of behavioural responses – acting on an object or perceiving it only.
This project aims to investigate differences that can be observed between lab experiments and actual AR applications, in order to provide a way to transpose AR research regarding basic perception issue from labs to ecological applications. We will develop an approach in which we take into account the real scene complexity (using natural scene statistics derived from the human vision literature) to adjust the appearance of digital elements in scene. We will test human performance for interacting with these objects using both perceptual and action based tasks.

Option 3

Immersive representation of interacting visualisations in Augmented Reality

Augmented Reality (AR) allows to present digital images, objects and environments superimposed on the real world. It can be used to render elements all around the user, and help him or her to interact with complex immersive data, such as robot swarms or procedure guidance augmentations.
How to represent such data is still a major challenge. Many elements can require the user’s attention as the same time, and the superimposition of many data in real environments, coupled with AR related issues such as incomplete occlusions, can easily overwhelm users. In addition, AR is known to increase workload and context switching between real and augmented elements can also create attention cost. Balancing these constraints could produce a new way of presenting data, but any solutions in this area are likely to be context specific. That is, it will be necessary to understand how to characterise and resolve conflicts between data visualisation and elements in the environment to enhance a user’s performance using the AR application.
This project aims to improve such 360° immersive visualizations while taking into account the perceptual specificities of AR. We will compare performance on immersive 360deg visualisations with more restricted views using a real-world scenario. Part of the contribution of this project will be the development of a protocol to assess how gains in data assessment can be related to the minimisation of attentional interference.

Prof. Gilles Coppin

Dr. Étienne Peillard

Prof. Anna Ma-Wyatt

Computer Science, Psychology