Quanxiao Zhang
Papers
1
Total Citations
7
H-Index
1
About
Quanxiao Zhang is at the forefront of energy-efficient 3D perception, pioneering the integration of spiking neural networks (SNNs) into point cloud processing. His landmark work, "Point-to-Spike Residual Learning for Energy-Efficient 3D Point Cloud Classification" (2024), introduces a novel framework that bridges the gap between biologically inspired neural computation and real-world 3D data. By designing a point-to-spike encoding scheme and residual learning architecture, Zhang enables SNNs to achieve competitive classification accuracy on point clouds while consuming dramatically less power than traditional deep networks. This breakthrough addresses a critical bottleneck in deploying AI on edge devices, where energy constraints are paramount. With 7 citations already in its first year, the paper signals growing recognition of his contributions to sustainable AI. Zhang’s research not only advances the theoretical understanding of spiking dynamics in non-Euclidean spaces but also opens practical pathways for low-power robotics, autonomous navigation, and augmented reality systems. His work stands as a vital step toward making 3D vision as energy-efficient as biological vision.
Research Focus
Key Achievements
Top Papers
- 1