Huizheng Wang
Papers
2
Total Citations
13
H-Index
2
About
Huizheng Wang is a rising star in the field of energy-efficient computing architectures, with a primary focus on **computation-in-memory (CIM) systems** for advanced **stereo vision processing**. Their most impactful work centers on the development of **CV-CIM**, a novel hardware accelerator that revolutionizes cost-volume construction—the critical kernel that computes pixel similarities between paired images for depth perception. Wang’s key contribution is the introduction of a **hybrid-domain XOR-derived similarity-aware architecture**, which dramatically reduces the energy and latency overheads traditionally associated with large parameter sizes and consecutive data accesses in stereo vision. This innovation directly supports demanding real-time applications in robotics, autonomous driving, 3D reconstruction, and AR/VR. Their 2023 paper on the 28nm CV-CIM chip has already garnered 10 citations, while a 2024 follow-up extends the architecture’s versatility. By tackling the fundamental bottleneck of data movement in vision pipelines, Wang is paving the way for more efficient, on-device spatial intelligence—a critical enabler for next-generation autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2