Shaheer Mohamed
Commonwealth Scientific and Industrial Research Organisation
Papers
1
Total Citations
30
H-Index
1
About
Shaheer Mohamed is a rising researcher at the forefront of hyperspectral imaging and deep learning, with a particular focus on advancing transformer architectures for complex remote sensing data. His most notable contribution, the <i>FactoFormer</i> model, introduces a novel factorized hyperspectral transformer that decouples spectral and spatial processing, enabling more efficient learning of long-range dependencies within high-dimensional imagery. By integrating self-supervised pretraining, Mohamed’s work addresses the critical challenge of limited labeled data in hyperspectral analysis, achieving superior performance in classification tasks. Though early in his career, his flagship paper has already garnered 30 citations, signaling strong impact and growing recognition within the computer vision and geoscience communities. Mohamed’s research bridges the gap between state-of-the-art natural language processing techniques and remote sensing applications, paving the way for more robust, data-efficient models. His innovative approach to factorizing transformer attention mechanisms stands out as a promising direction for scalable hyperspectral image understanding, making him a researcher to watch in the evolving landscape of AI-driven Earth observation.
Research Focus
Key Achievements
Top Papers
- 1