Mogomotsi Keaikitse
Papers
1
Total Citations
3
H-Index
1
About
Mogomotsi Keaikitse is a researcher whose work lies at the intersection of computer vision, robotics, and probabilistic modeling, with a particular focus on active object recognition. In his most-cited paper, "Probabilistic Active Recognition of Multiple Objects Using Hough-Based Geometric Matching Features" (2014), Keaikitse introduced a novel framework that combines Hough-based geometric matching with probabilistic reasoning to enable robots to actively and efficiently identify multiple objects in cluttered environments. This contribution is significant because it addresses a core challenge in autonomous systems: how to balance the cost of sensing with the need for accurate recognition in real-time. By leveraging geometric features and probabilistic inference, his approach allows for robust object detection even under partial occlusion or noisy sensor data. Although his citation count is modest—with the paper garnering three citations—the work demonstrates a thoughtful integration of classical computer vision techniques with modern probabilistic methods. Keaikitse’s research is particularly valuable for students and engineers working on robotic perception, offering a principled way to design active recognition systems that are both computationally efficient and reliable in dynamic settings.
Research Focus
Key Achievements
Top Papers
- 1