Majid Mirmehdi

Papers

Total Citations

H-Index

About

Majid Mirmehdi is a leading researcher in computer vision, with key contributions spanning text detection in natural scenes, assistive robotics, and document image analysis. His seminal work, "Finding Text Regions Using Localised Measures" (2000, 44 citations), introduced a pioneering method that leverages statistical properties of local image neighborhoods to locate text in real-world scenes—a foundational technique for applications in robot vision, wearable computing, and autonomous systems. This work remains highly influential in the field of scene text understanding. Mirmehdi has also advanced robotics through innovative approaches to kinematic modeling, including "Bootstrapping a robot’s kinematic model" (2013, 9 citations) and "Building a Kinematic Model of a Robot’s Arm with a Depth Camera" (2012, 5 citations), which enable robots to autonomously learn their own physical parameters using depth sensors. Additionally, his research on "A non-contact method of capturing low-resolution text for OCR" (2003, 8 citations) addresses practical challenges in document analysis. Mirmehdi’s work bridges fundamental computer vision theory with real-world applications, making him a respected figure in both academic and applied research communities.