RadarSense: Radar-based Sensing Algorithms, Techniques, and Applications for Novel Interactions with Computing Systems

   

Project summary

Project no.: PN-III-CEI-BIM-PBE-2020-0001; Contract no.: 1BM/2021
Principal Investigators: Radu-Daniel Vatavu (University of Suceava, Romania) and Jean Vanderdonckt (Université catholique de Louvain, Belgium)
Funded by UEFISCDI, Romania and Wallonie-Bruxelles International, Belgium
Funding scheme: PNIII P3 - European and International Cooperation
Running period: January 2021 - December 2022 (24 months)

Abstract

We plan to develop in this project new algorithmic approaches, software tools, and applications for gesture interaction based on radar sensing technology. To this end, we start by addressing the lack of structured knowledge in the community regarding radar-based algorithms and input techniques for gesture interaction, which we believe prevents thorough understanding of the practical opportunities of this technology. We propose a new algorithmic technique for recognizing radar gestures as well as applications of gesture-based interaction.

Objectives

The main objective of this project is to facilitate and support the scientific collaboration between Stefan cel Mare University of Suceava and Université catholique de Louvain to foster research activities on the topic of gesture-based interaction. Our specific objectives are:
  1. Perform a critical analysis of the input techniques and applications for radar-based interaction with computing systems
  2. Propose a new algorithmic approach for effective recognition of radar-based gestures
  3. Demonstrate radar-based input for practical interactive computer systems

Team

University of Suceava, Romania Université catholique de Louvain, Belgium
  •  Prof. Radu-Daniel Vatavu, Principal Investigator
  •  Cristian Pamparău, PhD student
  •  Alexandru-Ionut Șiean, PhD student
  •  Adrian-Vasile Catană, PhD student
  •  Laura-Bianca Bilius, PhD student
  •  Irina Popovici, PhD student
  •  Adrian Aiordăchioae, PhD student
  •  Prof. Jean Vanderdonckt, Principal Investigator
  •  Prof. Sébastien Lambot
  •  Laetitia Lambillotte, PhD student
  •  Nathan Magrofuoco, PhD student
  •  Soreangsey Kiv, PhD student
  •  Santiago Villarreal, PhD student
  •  Mehdi Ousmer, PhD student
  •  Quentin Sellier, PhD student
  •  Arthur Sluÿters, PhD student
  •  Sanae Abdelouassaa, Master's student
The research activities are conducted in the Machine Intelligence and Information Visualization (MintViz) Research Laboratory of the MANSiD Research Center (Romania) and the Louvain Interaction Laboratory (Belgium).

Publications

  1. Santiago Villarreal-Narvaez, Arthur Sluÿters, Jean Vanderdonckt, Radu-Daniel Vatavu. (2023). Brave New GES World: A Systematic Literature Review of Gestures and Referents in Gesture Elicitation Studies. ACM Computing Surveys (Dec. 2023). ACM, New York, NY, USA, 39 pages
    IF: 16.6 | 5-Year IF: 18
    PDF | DOI | Supplementary material PDF
  1. Stefano Chioccarello, Arthur Sluÿters, Alberto Testolin, Jean Vanderdonckt, Sébastien Lambot. (2023). Forte: Few Samples for Recognizing Hand Gestures with a Smartphone-attached Radar. Proceedings of the ACM on Human-Computer Interaction 7(EICS). ACM, New York, NY, USA, Article 179, 25 pages
    PDF | DOI
  1. Alexandru-Ionuț Șiean, Cristian Pamparău, Arthur Sluÿters, Radu-Daniel Vatavu, Jean Vanderdonckt. (2023). Flexible Gesture Input with Radars: Systematic Literature Review and Taxonomy of Radar Sensing Integration in Ambient Intelligence Environments. Journal of Ambient Intelligence and Humanized Computing. Springer, 15 pages
    IF: 3.662 | 5-Year IF: 3.718
    PDF | DOI
  1. Arthur Sluÿters, Sébastien Lambot, Jean Vanderdonckt, Radu-Daniel Vatavu. (2023). RadarSense: Accurate Recognition of Mid-Air Hand Gestures with Radar Sensing and Few Training Examples. ACM Transactions on Interactive Intelligent Systems. ACM, New York, NY, USA, 45 pages
    IF: 1.887 | 5-Year IF: 5.466
    PDF | DOI
  1. Arthur Sluÿters, Sébastien Lambot, Jean Vanderdonckt. (2022). Hand Gesture Recognition for an Off-the-Shelf Radar by Electromagnetic Modeling and Inversion. In Proceedings of IUI '22, the 27th International Conference on Intelligent User Interfaces, 506-522. ACM, New York, NY, USA
    ACCEPTANCE RATE: 24.5% (62/253) | ARC CORE A* |    BEST PAPER AWARD   
    PDF | DOI
  1. Alexandru-Ionuț Șiean, Cristian Pamparău, Radu-Daniel Vatavu. (2022). Scenario-based Exploration of Integrating Radar Sensing into Everyday Objects for Free-Hand Television Control. In Proceedings of IMX '22, the ACM International Conference on Interactive Media Experiences, 357-362. ACM, New York, NY, USA
    PDF | DOI
  1. Santiago Villarreal-Narvaez, Alexandru-Ionuţ Şiean, Arthur Sluÿters, Radu-Daniel Vatavu, Jean Vanderdonckt. (2022). Informing Future Gesture Elicitation Studies for Interactive Applications that Use Radar Sensing. In Proceedings of AVI '22, the International Conference on Advanced Visual Interfaces, 50:1-50:3. ACM, New York, NY, USA
    ARC CORE B | PDF | DOI | WOS:001051742000055

Project reports (in Romanian)

  1. Final report PDF
  2. Scientific report 2022 PDF
  3. Scientific report 2021 PDF

Media

The following videos demonstrate combined touch input and mid-air gesture interactions. Mid-air gestures are sensed by the Walabot radar that is placed on the lower frame of the TV set in the first video and under the table in the second video. Other locations for placing a radar sensor are possible as discussed in our paper (Siean et al., 2022) on this topic.



The following testimonial video features some of our team members talking about their research topics of interest in relation to this project, but also the larger context of the ongoing scientific collaboration between Université catholique de Louvain and Ștefan cel Mare University of Suceava. Video recording and post production was done at LouRIM (Louvain Research Institute in Management and Organizations) by Laura Vecchiato of the UCL Coordination of Doctoral Activities & Communication Support.

Software

Our software application implementing combined touch and mid-air gesture interactions can be downloaded here: RAR



Note: flag icons used on this page (flag-icons-css) are available under the MIT license here.