Expert to build privacy-safe AI for smart spaces

Nigerian doctoral candidate, James Onyejizu

Kindly share this story:

Amid global concerns about privacy through smart cameras and continuous data collection, a Nigerian-born engineer, James Onyejizu, has said he is pioneering privacy-preserving, AI-driven multimodal sensing systems.

He said his research is focused on building smart spaces that can sense human presence and activity without compromising personal privacy.

In an interview with Saturday PUNCH, Onyejizu, a doctoral candidate in Electrical, Computer, and Systems Engineering at Rensselaer Polytechnic Institute, New York, explained that his work would enable the estimation of human presence and activity in buildings without the use of invasive cameras or personally identifiable imagery.

Onyejizu said, “My project, ‘Privacy-preserving AI-driven multimodal sensing for human activity estimation,’ focuses on using non-imaging sensors to infer who is in a room, where they are, and what they are doing, all while keeping individuals’ identities protected.

“Most smart building systems today assume you must watch people to understand them. My work starts from the opposite belief: we can design systems that are intelligent and responsive to people while still being deeply respectful of their privacy.

“Using rich but privacy-preserving data, I am developing graph-based deep learning models that can detect occupancy and estimate how many people are in the room, infer basic activities (e.g., sitting, walking, presenting, group discussion), and connect these patterns to indoor air quality and energy use, enabling smarter ventilation and HVAC control.”

The engineer explained that his work sits at the core of a new vision for smart buildings, adding that they will be spaces that adapt to people in real time, improve health and comfort, and reduce energy waste.

Highlighting the importance of privacy-preserving activity sensing for Nigeria and the rest of the African continent, Onyejizu noted that it will enhance trust and adapt to culture.

“Many communities are understandably sceptical of heavy surveillance. Systems that use non-camera sensors to manage comfort and safety can build trust while still delivering the benefits of smart buildings.

“Electricity is often expensive and sometimes unstable. So, if schools, hospitals, and offices can automatically adjust ventilation and cooling based on real activity, not just fixed schedules, they can save significant energy while maintaining good indoor air quality.

“Just as African countries can leapfrog to mobile banking and distributed solar, they can leapfrog to privacy-respecting smart buildings rather than importing camera-heavy systems developed elsewhere. I hope that the methods I am developing—graph-based models, multimodal sensing, and privacy-aware algorithms—can be adapted to low-cost sensors and deployed in classrooms, clinics, and offices across the continent,” he added.

Kindly share this story:

All rights reserved. This material, and other digital content on this website, may not be reproduced, published, broadcast, rewritten or redistributed in whole or in part without prior express written permission from PUNCH.

Contact: [email protected]

UPDATE NEWS: Nigerians can now invest ₦2.5million on premium domains and profit about ₦17-₦25Million. All earnings paid in US Dollars. Rather than wonder, 👉 click here to find out how it work (Be sure to ask for evidence of clients that have benefitted from this)

Stay informed and ahead of the curve! Follow The Punch Newspaper on WhatsApp for real-time updates, breaking news, and exclusive content. Don't miss a headline – join now!

Stay in the know—fast. Get instant alerts, breaking headlines, and exclusive stories with the Punch News App. Download now and never miss a beat.