Md Awsafur Rahman
Papers
2
Total Citations
16
H-Index
2
About
Md Awsafur Rahman is a rising computer vision researcher whose work focuses on advancing monocular depth estimation—a critical technology for autonomous driving, robotics, and augmented reality. His most notable contribution, **DwinFormer** (2023, 11 citations), introduces a novel dual-window transformer architecture that overcomes the longstanding trade-off between global consistency and fine-grained local detail in depth maps, achieving state-of-the-art end-to-end performance. Building on this, Rahman’s **semi-supervised framework** (2024, 5 citations) integrates semantic and depth information through a symbiotic transformer and the innovative NearFarMix augmentation strategy, addressing the scarcity of labeled semantic data in real-world datasets. This work demonstrates his ability to tackle practical data limitations while enhancing scene understanding. Though early in his career, Rahman’s publications have already garnered significant attention for their technical elegance and real-world applicability. His research sits at the intersection of transformer architectures, multi-task learning, and data-efficient AI—promising to make autonomous systems safer and more perceptive. With a clear trajectory toward solving fundamental perception challenges, Rahman is a researcher to watch in the coming years.
Research Focus
Key Achievements
Top Papers
- 1
- 2