Thursday, Novembrer 13th, 2025
- Share
- Share on Facebook
- Share on X
- Share on LinkedIn
Representation of interactions based on hand microgestures
Abstract
Microgestures are quick and subtle finger movements. They are becoming increasingly popular and appear particularly promising for mobile interaction with wearable devices such as smartwatches and Augmented Reality (AR) headsets. Previous research has mostly focused on what they can be used for, how to detect them and on which microgestures are relevant according to the end-users’ preferences. Whether in research papers or in the latest Apple Watch features, there are images meant to represent how to perform microgestures. These representations are the most prominent way to present microgestures to the end-users but they often remain disparate and ambiguous. Yet, it is still unclear what is characteristic of a “good” representation of a microgesture and how such representations could be used to create usable help intefaces with more than just 4 or 5 microgestures. In this thesis, we propose a two-part approach: (1) determining how to represent the microgestures that can be performed and (2) designing a help inteface for an application using multiple commands associated to microgestures. In the first part, we begin by reviewing the related work. In doing so, we determine a representative set of microgestures (hand-free tap, swipe, flex and hold), classify the existing visual cues used in their representations and explore how help interfaces are usually evaluated in Human-Computer Interaction. Then, we investigate the fundamental aspects of single-picture representations of microgestures, i.e. representations with one microgesture per hand shape. For those, we propose 21 “families”, i.e. groups of representations sharing a common design, and we compare them in two complementary experiments. The first is an online form that allow us to rank families and determine which ones work best for each type of microgesture. The second, is a laboratory experiment with an AR headset that allow us to gather qualitative data on the most promising families. With these results combined, we provide a set of design guidelines and a metric to evaluate the explicitness of a single-picture representation of microgestures. In the second part, we first compare different design strategies that can be used to conceive simultaneous representations of microgestures, i.e. representations that show multiple microgestures on the same hand shape. We then design a simultaneous representation of microgestures with their associated commands for a music player application and test its usability in a Wizard of Oz experiment. We also qualitatively compare its perfomance on small screens, namely smartwartches and smartphones, with a mosaic of single-picture representations of microgestures. With the results of these three experiments, we provide design recommentations specifically thought for simultaneous representations of microgestures. At this point, we are able to say that even though single-picture representations of microgestures are quicker to understand, simultaneous representations of microgestures can still be understood and used by complete novice users after only a few seconds. Based on the results of the previous chapters, we then conceive and compare 3 help intefaces designed for a smartwatch application. This last experiment allows us to determine the advantages and drawbacks of each help inteface and suggest that help interfaces for smartwatch applications should use a swipeable carousel of simultaneous representations of microgestures to foster both the discoverability and learning of the available commands. We also propose a first formalization of the confusion between two microgestures. Finally, we present a Unity package for AR applications and a Python package that both allow to create representations of microgestures.
Date and place
Thursday, Novembrer 13th at 13:00
Salle de Séminaire 1, Bâtiment IMAG
Supervision
Laurence Nigay, Sylvain Malacria and Alix Goguey
IHM Team
Jury members
Laurence Nigay
Professeure des Universités, Université Grenoble Alpes, Thesis Co-director
Sylvain Malacria
Chargé de Recherche, Thesis Co-director
Alix Goguey
Maître de conférences, Université Grenoble Alpes, Thesis Supervisor
Adwait Sharma
Lecturer, University of Bath, Examiner
Dominique Vaufreydaz
Professeur des Universités, Université Grenoble Alpes, Examiner
Katrin Wolf
Professor, Berlin University of Applied Sciences and Technology, Examiner
Caroline Appert
Directrice de recherche, CNRS, Reporter
Radu-Daniel Vatavu
Professor, "Ștefan cel Mare" University of Suceava, Reporter
- Share
- Share on Facebook
- Share on X
- Share on LinkedIn