Fuseini Mumuni
Papers
5
Total Citations
118
H-Index
5
About
Fuseini Mumuni is a researcher at the forefront of computer vision and deep learning, with a focus on making autonomous systems more perceptive, robust, and interpretable. His work spans geometric deep learning, explainable AI, and robot navigation, where he addresses fundamental challenges in how machines understand and interact with dynamic environments. Mumuni’s most influential work, a comprehensive review of CNN architectures for geometric transformation-invariant feature representation (63 citations), has become a key reference for researchers tackling viewpoint and deformation invariance in visual recognition. He has also made significant contributions to enhancing deep learning’s reliability, proposing frameworks that integrate prior knowledge and cognitive models to improve explainability, adversarial robustness, and zero-shot learning (21 citations). In applied robotics, Mumuni has advanced UAV navigation by developing deep learning methods for monocular depth estimation, optical flow, and ego-motion with geometric guidance (15 citations), as well as Bayesian cue integration for fusing structure-from-motion with CNN-based depth predictions (12 citations). His survey on robust appearance modeling for object detection and tracking (7 citations) further underscores his impact. With over 118 total citations, Mumuni’s work bridges theory and practice, offering elegant solutions for safer, more intelligent autonomous navigation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5