Vijaykrishnan Narayanan
Papers
6
Total Citations
82
H-Index
5
About
Vijaykrishnan Narayanan is a researcher whose work spans neuromorphic computing, computer vision, and multimodal depth estimation, with a particular focus on developing efficient and biologically inspired hardware and algorithmic solutions for real-world perception tasks. His early contributions include pioneering work on reconfigurable accelerators for neuromorphic object recognition, drawing on biologically motivated models such as HMAX to advance machine vision beyond conventional approaches. He has also explored the human visual system's multi-resolution properties, proposing saliency-driven foveation frameworks that mimic how biological systems selectively attend to regions of interest. More recently, Narayanan has made significant strides in multimodal depth estimation, combining event cameras with traditional RGB sensors and leveraging transformer-based architectures and generative adversarial networks to achieve robust, pixel-accurate depth prediction for autonomous navigation and robotics. His work on sparse-to-dense depth completion and hybrid spiking neural network–ANN architectures reflects a commitment to embedded and energy-efficient vision systems. With citations spanning foundational neuromorphic hardware to cutting-edge fusion models, his research collectively bridges neuroscience-inspired computing and practical deployment in autonomous systems, making his body of work highly relevant to students and researchers working at the intersection of embedded vision, deep learning, and robotics.
Research Focus
Key Achievements
Top Papers
- 1A reconfigurable accelerator for neuromorphic object recognition27 citations · 2012
- 2A multi-resolution saliency framework to drive foveation19 citations · 2013
- 3
- 4
- 5
- 6