Guozhen Zhu
Principal Scientist @ Origin AI
gzzhu [at] terpmail.umd.edu
Hi, I’m Guozhen Zhu, currently working with Dr. K. J. Ray Liu at Origin AI, where I lead cutting-edge research in context-aware AI systems and ambient intelligent sensing technologies. I received my M.S. and Ph.D. in Electrical Engineering from the University of Maryland, College Park, in 2022 and 2023, respectively.
My research aims to make intelligent systems more aware, adaptive, and trustworthy in the real world. I focus on enabling devices to sense, understand, and respond to human activity and environmental context in a contactless, scalable, and privacy-preserving way—empowering applications in smart homes, healthcare, security, and beyond.
My current research focuses on building robust and generalizable intelligence from real-world ambient signals:
- Signal-driven representation learning - extracting interpretable and resilient features that disentangle target-relevant information from noisy, entangled spatiotemporal signals.
- Edge AI - designing efficient models for real-time inference and learning on resource-constrained edge devices.
- Generalizable intelligence – developing methods that address distribution shifts and variability across users, devices, and environments to ensure reliable performance in real-world conditions.
I like cool and artistic research works but believe in those solving real-world problems.
Outside of work, you’ll often find me on the tennis court
, playing the piano
, baking something semi-sweet
, or soaking in nature’s beauty
.
Thanks for stopping by! I’m always open to discussions, new ideas, and meaningful collaborations—don’t hesitate to reach out!
News
| Sep 18, 2025 | 🎉 Our paper CSI-Bench has been accepted by NeurIPS 2025! I will be presenting the paper in San Diego. |
|---|---|
| Jun 15, 2025 | 📝 Presented key insights and lessons learned from deploying a large-scale WiFi sensing system, detailed in a peer-reviewed paper. |
| Jun 01, 2025 | 🎉 We released CSI-Bench, the first large-scale, in-the-wild benchmark dataset for multi-task WiFi sensing. Now fully open-sourced to accelerate research in ambient intelligence and edge AI! |
| Mar 01, 2025 | 🏆 A patent granted for “Method, Apparatus, and System for Wireless Proximity Detection,” enabling motion sensing on over 100 million Signify Wiz Bulbs. |
| Jan 01, 2025 | 🏅 Our product TruShield earned the 2025 CES Twice Picks Award, recognizing innovation in contactless human presence detection and privacy-preserving sensing. |
| Sep 01, 2024 | 📋 Patent granted for “Map Reconstruction Based on Wireless Tracking.” |
Industry Impact
Technology deployed worldwide across multiple sectors:
- Verizon Home Awareness (2022)
WiFi sensing deployed on more than 10 million Fios routers nationwide
Based on US Patent No. US12256360B2 and App. No. 18/391,529. - Signify (Philips) & Nami Sensing (2025)
Proximity detection and motion sensing deployed on more than 100 million smart lights and plugs
Based on US Patent No. US12256360B2, and App. No. 18/391,529. - TruShield Security (2025)
Award-winning privacy-preserving security system (CES 2025 TWICE Picks Award)
Based on US Patent No. US12352889B2 and App. No. 18/391,529. - Verisure WiFi Vision & Deutsche Telekom (2024)
EU security deployment with Internet Service Providers
Based on EU Patent App. No. 24150597.3 and US Patent App. No. 18/391,529. - Arlo Technologies (2024)
Exclusive partnership for Verified Human Presence detection
Based on US Patent No. US12352889B2 and App. No. 18/391,529.