Robots might be getting smarter, but to truly support people in daily life, they also need to become more empathetic. That means recognizing and responding to human emotions in real time.
Most facial recognition models are trained once and then expected to work across every scenario. However, a model trained on one dataset often struggles when faced with new situations, and retraining from scratch is slow and inefficient.
Dr. Rahul Singh Maharjan and his team are tackling this challenge by developing a new approach: teaching AI to learn emotions incrementally. Instead of forgetting what it already knows, the system builds on past experiences while adapting to fresh emotional data. This makes it more resilient and better prepared for real-world human interaction.
As Dr. Maharjan explains, “For technology to truly integrate into our lives, it must understand our emotions. My goal is to help build AI that doesn’t just compute, but connects with us.”
His most recent study is published in the journal IEEE Transactions on Affective Computing.
More information:
Rahul Singh Maharjan et al, Continual Facial Features Transfer for Facial Expression Recognition, IEEE Transactions on Affective Computing (2025). DOI: 10.1109/taffc.2025.3561139
Citation:
Creating robots that adapt to your emotion (2025, September 23)
retrieved 24 September 2025
from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
