Kaijie Wei
Papers
2
Total Citations
7
H-Index
2
About
Kaijie Wei is a researcher advancing the frontier of real-time robot audition through hardware-software co-design. His core research focuses on low-power, high-performance implementations of sound source localization and separation—critical functions that allow robots to hear and understand their environments. Wei’s most notable contributions involve porting modules from HARK, the open-source “OpenCV for audio,” onto FPGA boards. His 2023 paper on a low-power implementation of Geometric High-order Decorrelation-based Source Separation achieved 4 citations, while his 2022 work on FPGA off-loading of HARK sound source localization garnered 3 citations. Though early in his citation trajectory, these works address a fundamental bottleneck: making computationally intensive auditory processing feasible for power-constrained robotic platforms. By off-loading the foundational task of sound source localization to reconfigurable hardware, Wei enables faster, more energy-efficient robot audition—paving the way for truly responsive autonomous systems. His work stands at the intersection of embedded systems, signal processing, and robotics, offering practical solutions for deploying sophisticated auditory perception in the real world.
Research Focus
Key Achievements
Top Papers
- 1
- 2An FPGA off-loading of HARK sound source localization3 citations · 2022