Pramod Swami
Papers
2
Total Citations
11
H-Index
2
About
Pramod Swami is a researcher at the forefront of efficient deep learning and computer vision systems for embedded and heterogeneous platforms. His work focuses on bridging the gap between high-performance neural networks and the constraints of real-world, resource-limited devices. Swami’s key contributions lie in developing novel quantization and software frameworks that enable the practical deployment of complex models. His highly cited paper, “CNN Inference: Dynamic and Predictive Quantization” (2018, 6 citations), addresses the critical challenge of minimizing accuracy loss while compressing Convolutional Neural Networks for embedded use in automotive, medical, and robotics applications. Complementing this, his work on a “Novel OpenVX implementation for heterogeneous multi-core systems” (2017, 5 citations) provides a software framework to maximize utilization of diverse computing elements like CPUs, GPUs, and DSPs, ensuring low latency for computer vision tasks in autonomous systems and AR/VR. Through these contributions, Swami is enabling a new generation of intelligent, edge-based devices, making him a key figure in the practical advancement of embedded AI.
Research Focus
Key Achievements
Top Papers
- 1CNN Inference: Dynamic and Predictive Quantization6 citations · 2018
- 2Novel OpenVX implementation for heterogeneous multi-core systems5 citations · 2017