Jiacong Sun
Papers
1
Total Citations
1
H-Index
1
About
Jiacong Sun is a leading researcher in energy-efficient hardware acceleration for point-cloud neural networks (PNNs), with a focus on three-dimensional (3D) deep learning for autonomous driving, robotics, and virtual reality. His most-cited work, "Nebula: A 28nm 109.8TOPS/W 3D PNN Accelerator Featuring Adaptive Partition, Multi-Skipping, and Block-Wise Aggregation," introduces a groundbreaking chip design that achieves exceptional energy efficiency—109.8 tera-operations per second per watt—by leveraging adaptive partitioning and multi-skipping techniques to reduce redundant computations in point-cloud processing. This work addresses the critical challenge of deploying point-based PNNs in real-time, resource-constrained environments, where their superior accuracy often comes at the cost of high computational demand. Sun’s contributions have been recognized for advancing the practical viability of 3D perception systems, with his accelerator design cited as a key reference in the field. His research bridges the gap between algorithmic innovation and hardware implementation, enabling more efficient and scalable AI for spatial understanding.
Research Focus
Key Achievements
Top Papers
- 1