Gedamu Alemu Kumie

University of Electronic Science and Technology of China

Papers

Total Citations

H-Index

About

Gedamu Alemu Kumie is a researcher at the forefront of computer vision, with a primary focus on view-invariant action recognition and human–robot interaction. His most-cited work, "Dual-attention Network for View-invariant Action Recognition" (2023, 13 citations), tackles a fundamental challenge in the field: accurately recognizing human actions despite occlusions and information loss caused by changing camera perspectives. By introducing a novel dual-attention mechanism, Kumie’s model effectively captures spatial and temporal dependencies, enabling robust performance across varying viewpoints. This contribution is critical for real-world applications like visual surveillance and intuitive human–robot collaboration. Beyond this flagship paper, his research portfolio demonstrates a sustained commitment to advancing deep learning architectures for action understanding, with a focus on mitigating viewpoint-induced distortions. Kumie’s work has garnered attention for its practical relevance and technical innovation, offering a pathway toward more resilient and adaptive recognition systems. His achievements underscore a dedication to solving complex, real-world vision problems, making him a notable voice in the ongoing evolution of action recognition technology.