Zhetong Huang

About

Zhetong Huang is a rising researcher in the field of efficient deep learning acceleration, with a primary focus on hardware-software co-design for 3D perception systems. His work bridges the critical gap between advanced computer vision algorithms and practical hardware implementation, particularly for autonomous driving and robotics applications. Huang’s most cited work, “VEA: An FPGA-Based Voxel Encoding Accelerator for 3D Object Detection with LiDAR” (2022, 11 citations), addresses a fundamental bottleneck in real-time 3D perception: the computational inefficiency of processing sparse and unstructured point cloud data. By designing a specialized FPGA accelerator for voxel encoding—a key preprocessing step in many LiDAR-based detection pipelines—Huang demonstrated how custom hardware can dramatically improve throughput while maintaining detection accuracy. This contribution is particularly significant as autonomous systems demand both high performance and low latency under strict power constraints. Though early in his career, Huang’s work exemplifies the growing importance of domain-specific accelerators in enabling next-generation perception systems, and his research direction holds promise for making 3D object detection more practical for real-world deployment in vehicles and robots.

Research Focus

Key Achievements

Top Papers

1. 1
   VEA: An FPGA-Based Voxel Encoding Accelerator for 3D Object Detection with LiDAR

   11 citations · 2022

Key Collaborators

Contact & Links