Vidya Kamath
Papers
3
Total Citations
23
H-Index
3
About
Vidya Kamath is a researcher at the forefront of embedded computer vision and edge AI, specializing in deploying deep learning models on resource-constrained hardware. Her work addresses a critical challenge in robotics and IoT: enabling real-time object detection and tracking using low-cost, low-power devices like the Raspberry Pi and NVIDIA Jetson Nano. Kamath’s most-cited paper (10 citations) introduces an exploratory framework for data preparation and model training tailored to Raspberry Pi-based deployments, bridging the gap between high-quality pretrained models and the noisy, low-resolution inputs from real-world cameras. She further advances human-robot interaction by investigating MobileNet-SSD for autonomous human-following robots, achieving robust stand-alone tracking (9 citations). Her analysis of lightweight models on Jetson Nano using TensorFlow Lite (4 citations) provides practical guidelines for balancing accuracy and latency in real-time applications. Kamath’s contributions are pivotal for democratizing AI, making sophisticated computer vision accessible to hobbyists, startups, and researchers working with limited hardware. Her work not only pushes the boundaries of edge intelligence but also offers reproducible methodologies that accelerate deployment in autonomous systems, smart surveillance, and assistive robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3