Yakun Wu
Papers
2
Total Citations
23
H-Index
2
About
Yakun Wu is a researcher at the forefront of embedded artificial intelligence, specializing in the hardware acceleration of computer vision algorithms on FPGA platforms. Their work directly addresses the critical challenge of deploying computationally intensive deep neural networks—such as those used for visual semantic segmentation and Simultaneous Localization and Mapping (SLAM)—onto resource-constrained edge devices. Wu’s major contributions include the design of EDSSA, an encoder-decoder semantic segmentation network accelerator that leverages OpenCL-based FPGA technology to achieve high performance for applications in autonomous driving and intelligent robotics. Building on this, Wu developed an energy-efficient FPGA accelerator for DS-SLAM, a semantic SLAM system that enables mobile robots to robustly navigate dynamic environments. This work is pivotal for real-world robotics, where power efficiency and real-time processing are paramount. With their most cited papers accumulating over 20 citations, Wu is establishing a strong reputation for bridging the gap between sophisticated AI models and practical, low-power hardware implementations, making advanced robotic perception more accessible and deployable.
Research Focus
Key Achievements
Top Papers
- 1
- 2