Leibo Liu
Papers
2
Total Citations
12
H-Index
2
About
Leibo Liu is a leading researcher in energy-efficient hardware design for intelligent systems, with key contributions spanning computation-in-memory (CIM) architectures, approximate computing, and neuromorphic hardware. His most cited work, "7.7 CV-CIM: A 28nm XOR-Derived Similarity-Aware Computation-in-Memory for Cost-Volume Construction" (2023, 10 citations), introduces a novel CIM accelerator that dramatically improves the efficiency of stereo vision processing—a critical kernel for robotics, autonomous driving, and AR/VR applications. By enabling accurate pixel similarity computations with reduced data movement, this work addresses a major bottleneck in real-time 3D perception. Liu also pioneered the use of approximate circuits in hardware design for cerebellar models (2017), demonstrating how controlled computational imprecision can enable efficient motor control and image stabilization in robots. His research bridges the gap between biological inspiration and practical VLSI implementation, showing that complex neural models can be realized in compact, low-power silicon. With a focus on making advanced AI and vision algorithms deployable at the edge, Liu's work has direct implications for next-generation autonomous systems and smart sensors.
Research Focus
Key Achievements
Top Papers
- 1
- 2