Abinash Mohanty
Papers
1
Total Citations
16
H-Index
1
About
Abinash Mohanty is a leading researcher at the intersection of computer vision and reconfigurable hardware, specializing in efficient deep learning acceleration. His most impactful work focuses on end-to-end FPGA-based object detection, where he pioneered a pipelined CNN architecture integrated with non-maximum suppression—a critical contribution for real-time applications in autonomous driving, smart surveillance, and robotics. His 2021 paper on this topic has garnered 16 citations, demonstrating its influence in the field. Mohanty’s major contribution lies in bridging the gap between algorithmic complexity and hardware efficiency, enabling single-shot detectors (SSD) to run with low latency and high throughput on resource-constrained platforms. By optimizing the entire detection pipeline—from feature extraction to bounding box refinement—on FPGA fabric, he has advanced the practicality of embedded vision systems. His work is particularly notable for addressing the bottleneck of post-processing in object detection, making it viable for edge deployment. With a focus on end-to-end system design, Mohanty continues to shape how deep neural networks are deployed in real-world, latency-sensitive environments, inspiring both academic research and industrial innovation.
Research Focus
Key Achievements
Top Papers
- 1