Computer vision
"Computers are able to see, hear and learn. Welcome to the Future" - Dave Waters
"Work is demanding, but at the same time environment is friendly and relaxed. The professors are approachable and sincerely want you to succeed."
Sherif Mohamed
PhD researcher
Visual Odometry
We develop algorithms to analyze images captured by cameras to obtain the position and orientation of robots/vehicles.
Event Cameras
Event-based cameras, offer high potential to overcome the challenging task of detecting consistent corners in complex scenarios, such as high speed movements, low-lighting and High Dynamic Range (HDR) scenes. We design methods to unlock the capabilities of event cameras.
Dynamic Resource-aware
We develop methods to continuously monitor the H.W units of the embedded system and control the resource to maintain high real-time performance of complex algorithms on resource contained systems, such as TX2 and Raspberry pi.
How we analyze our algorithms
Real-time
Consistency
Accuracy
Robutsness
Efficiency
