Feras Dayoub

Dr. Feras Dayoub is a Senior Lecturer and Chief Investigator at the QUT Centre for Robotics (QCR), where he co-leads the Visual Learning and Understanding program. He previously served as Chief Investigator at the ARC Centre of Excellence for Robotic Vision (2016-2020) and holds a PhD in Robotics and Computer Vision from the University of Lincoln.

• Research Focus: Deploying computer vision and machine learning for real-world mobile robotics applications, including autonomous weed control (CRC-P), vision-enabled underwater robots for reef protection (COTSbot/RangerBot), and UAV-based infrastructure inspection.
• Teaching: Coordinates advanced robotics topics (EGB439) and teaches microprocessor systems (CAB202).

His recent publications focus on uncertainty quantification in robotic vision, open-set recognition, and performance monitoring of deployed models. His work has been recognized with multiple awards from the Australian Centre for Robotic Vision and a Google Impact Challenge popular vote award.

• 2023: Hyperdimensional Feature Fusion for Out-of-Distribution Detection
• 2023: Class Distribution Shift Prediction for Domain Adaptation
• 2022: Uncertainty for Open-Set Error Identification
• 2022: FSNet for Semantic Segmentation Failure Detection

Awards include:

• 2020: ACRV Best-Profile Raising Event
• 2019: QUT STEM Camp Certificate of Appreciation
• 2016: Google Impact Challenge People's Choice Award
• 2015: QUT Vice Chancellor's Performance Award

As supervisor, he guides projects on robotic object detection, cross-view localization, and continues to advance visual learning methodologies for real-world autonomous systems.