Papers
1
Total Citations
4
H-Index
1
About
Sukbin Lim is a leading researcher in embedded vision systems and FPGA-based acceleration for autonomous mobile robotics. His most impactful work centers on developing efficient hardware architectures that bridge the gap between high-performance convolutional neural networks (CNNs) and resource-constrained edge devices. Lim’s major contribution, the ACane platform, introduces an innovative “accumulation-as-convolution packing” technique that dramatically improves DSP utilization on FPGAs. This method enables autonomous robots to run complex vision models in real-time without sacrificing accuracy, addressing a critical bottleneck in mobile robotics. His 2024 paper on ACane has already garnered 4 citations, signaling growing recognition in the embedded systems community. Lim’s research is particularly notable for its practical impact—his designs are tailored for real-world deployment on autonomous platforms, where power efficiency and low latency are paramount. By optimizing low-precision quantization and DSP packing, he has demonstrated how FPGAs can rival GPU performance in vision tasks while consuming a fraction of the energy. For students and researchers exploring edge AI, Lim’s work offers a compelling blueprint for making deep learning truly mobile and autonomous.
Research Focus
Key Achievements
Top Papers
- 1