Keynote Speakers

Emotion-Aware Interfaces: Bridging Human and Intelligent Machines

Emotions are fundamental to human communication and play a critical role in shaping interactions. As Artificial Intelligence advances, there is a growing demand for systems that can not only process information but also understand and respond to human emotions. Emotion-aware interfaces represent an essential step toward this goal, enabling machines to recognize users’ affective states and adapt their responses in more natural and meaningful ways.

This talk introduces emotion-aware interfaces and outlines their potential impact across diverse domains, including human–robot collaboration, conversational agents, healthcare, customer engagement, and security. Our research contributions in areas such as facial (micro-)expression recognition, emotional gesture analysis, and remote heart rate estimation from video are presented with the discussion about emerging opportunities and key challenges that must be addressed to fully realize the promise of emotionally intelligent human–machine interaction.

Techno, Art, and Music Robots

With a background in electrical engineering and a passion for hands-on sound creation, Geist’s work is driven by a desire to interact physically with music. His robotic instruments are crafted using advanced technologies such as 3D printing, CNC milling, and laser cutting and have been shown all around the world.

In this talk, Geist will give insight into his art practice, share how he stopped working with human musicians and started working with music robots, and explain why AI music robots will not replace human musicians (soon).

Integrated Radar Sensing and Communications: convergence, co-design and learning

Integrated sensing and communications (ISAC) systems operate within shared, congested, or even contested spectrum, aiming to deliver high performance in both wireless communications and radio frequency (RF) sensing. Communications and sensing functionalities are co-designed for mutual benefit. ISAC systems share hardware and antenna resources, use joint waveforms, and function within shared spectra and exchange awareness about their radio environments to jointly optimize performance. ISAC technology has opened genuinely new lines of research and development in emerging 6G systems rather than merely being an evolution of 5G. ISAC takes advantage of the ongoing parallel convergence of modern radar and multi-antenna communications technologies. Sensing may be a service, can aid communications, and conversely, communications may aid sensing. ISAC is expected to provide unprecedented quality of experience towards reliable, resilient, and ubiquitous connectivity, beyond the high data-rate transmission of bits. This talk focuses on various waveform design, signal processing, optimization, adaptation and reinforcement learning techniques employed to achieve desired sensing and communications performances in multicarrier (MC) and multiantenna ISAC systems. Examples on waveform design, optimization, resource allocation in multi-user and multi target scenarios are provided.

From WiFi Sensing to Quantum Sensing: Toward a Ubiquitous Sensing Theory

WiFi/4G/5G based wireless sensing has attracted a lot of attention from both academia and industry in the last decade. However, most of the work focused on developing effective techniques for a certain application, very few work attempted to explore the fundamental sensing theory and answer fundamental questions such as the sensing model, sensing limit, sensing boundary and sensing quality of WiFi/4G/5G signals. In this talk, I will first introduce the Fresnel zone model we proposed in UbiComp 2016 as a generic theoretic basis for contactless human sensing with WiFi/4G/5G signals, revealing the relationship among the received CSI signal, the distance between the two transceivers, the location and heading of the sensing target with respect to the transceivers, and the environment. Then I will present the Sensing Signal to Noise Ratio (SSNR) as a new metric to inform the sensing limit, sensing boundary and sensing signal quality of WiFi/4G/5G-based human sensing systems. In order to further increase SSNR and push the wireless sensing limit, we explore the Rydberg theory about how the Rydberg atom interacts with RF signals and develop the world first Rydberg quantum sensing system which senses tiny variations of RF signals caused by human activities, showing significant performance increase compared to WiFi or mmWave Radar based sensing systems.

What will Wi-Fi 8 be? A primer on next generation Wi-Fi for communication and sensing

Wi-Fi technology has been traditionally focused on providing higher data rates through its standards and evolutions. However, recent applications and use cases have been pushing for highly reliable communications, demanding Wi-Fi to accommodate such requirements starting with Wi-Fi 8. This keynote presents an overview of Wi-Fi 8, highlighting the necessary changes and advancements brought by the standard to provide Ultra High Reliability (UHR) communication features. Besides introducing the new features, we will also discuss the increasingly important role of complementary functionalities, like Wi-Fi in sensing, focusing on the techniques enabling Access Points (APs) and User Equipments (UEs) to initiate and/or carry out sensing procedures. Finally, we will conclude by presenting research questions and open challenges to truly deliver UHR communications and to enhance sensing functionalities in Wi-Fi networks.  

The Trajectory of Ultrasonic Haptics: From Its History to Healthcare Applications, and Its Influence on Acoustic Levitation

Ultrasonic haptics has emerged as a powerful modality in mid-air tactile feedback, enabling contactless interactions that enrich immersive VR/AR experiences. This keynote traces its trajectory from its early conceptualization to the development of practical phased-array systems, culminating in real-world applications in healthcare. The talk will further explore how these techniques have influenced adjacent domains, particularly acoustic levitation, where precise wave control enables dynamic object manipulation in mid-air. By bridging fundamental physics with user-centered design, ultrasonic haptics demonstrates how wave-based modalities can serve as a versatile interface layer in multisensory environments. This journey offers not only a retrospective of technological evolution, but also a prospective look at cross-modal fusion and societal impact, providing insights for researchers aiming to integrate novel modalities into next-generation XR systems.

Assistive AR: Bridging Human Sensing and Everyday Assistance

Augmented Reality (AR) glasses have evolved from mere entertainment gadgets to practical wearable devices that enhance everyday life. This talk examines how AR can transform the way we interact with others and perceive our environment. Dr. Yunqi Guo will introduce Assistive AR—a class of AR systems his team is developing to support daily activities by improving both human–human and human–environment interactions. These systems combine human-like sensing capabilities, efficient cross-device computation, and intuitive generative displays. He will highlight their work on two key applications: breaking communication barriers for Deaf users through real-time, bidirectional sign language translation with AR glasses, and enhancing social interactions with proactive large language models (LLMs) that understand conversational context and offer response suggestions. Through AnySign—a team Dr. Guo founded during his PhD—this research is extending beyond academia to real-world assistance for end users. Looking ahead, AR’s ability to align closely with human sensing will enable a new generation of versatile, personalized assistance tools that seamlessly integrate into daily life.

From foundation models to tiny ML – the path ahead for ambient neuropsychiatry

Data is the new oil. AI is the new electricity. Every American will be wearing a wearable within four years. The clichés surrounding AI are endless. Perhaps there is some partial truth in them, but the real question is whether this suddenly fashionable tool will be useful and cost-effective. In the domain of health, we will need to overcome the barriers inherent in analyzing human physiology and activity, including distrust, noncompliance, laziness, misleading labels, misaligned incentives, limited resources, and climate armageddon. That seems like a lot to overcome, but I’m going to have a stab at outlining some promising pathways forward with a focus on neuropsychiatry.

Wearables 101: What Changes (and Doesn’t) as Your Data Scales

Every startup has many stories. This talk tells Oura’s through its data and the algorithms built on it through my lens. I’ll show how scientific and product-facing approaches meet, and how evidence thresholds, ground truth, and validation shift when the target is user value rather than publication. We’ll look at what changes as signals, users, and expectations grow, and what doesn’t. The aim is a practical lens on scaling ring-based data through algorithm highlights grounded in real user stories. I’ll also share what I wouldn’t change through principles that held up at scale to the collaboration that make it possible.