Ajay Kuzhively
Papers
1
Total Citations
16
H-Index
1
About
Ajay Kuzhively is a researcher at the forefront of efficient computer vision, specializing in hardware-accelerated deep learning for real-time object detection. His work bridges the critical gap between algorithmic performance and practical deployment, particularly in resource-constrained environments like autonomous driving, smart surveillance, and robotics. Kuzhively’s most notable contribution is his pioneering end-to-end FPGA-based object detection system, which integrates a pipelined convolutional neural network (CNN) with non-maximum suppression directly on hardware. This approach, detailed in his highly cited 2021 paper (16 citations), demonstrates how to achieve high-speed, low-latency inference for single-shot detectors (SSD) without sacrificing accuracy. By eliminating the traditional CPU-GPU bottleneck, his design enables robust, real-time classification and localization of objects directly on edge devices. This work has significant implications for embedded vision systems, offering a power-efficient alternative for applications where millisecond-level response times are critical. Kuzhively’s research continues to push the boundaries of efficient AI, making advanced computer vision accessible for practical, real-world deployment.
Research Focus
Key Achievements
Top Papers
- 1