Papers
3
Total Citations
57
H-Index
2
About
Kunyu Wang is a rising researcher at the intersection of medical imaging, embodied AI, and robotics. His work is defined by a commitment to enhancing machine perception and autonomous decision-making in real-world environments. Wang’s most impactful contribution to date is in medical computer vision, where he developed a novel approach to polyp detection that integrates advanced image pre-processing with an enhanced Faster R-CNN architecture. This work, published in 2020 and accumulating 54 citations, directly addresses the critical challenge of reducing missed polyps during colonoscopy—a procedure where the miss rate can reach 10%—thereby improving early diagnosis of colon cancer. More recently, Wang has pushed the boundaries of Embodied AI with NaVid, a video-based Vision-Language Model (VLM) designed to plan the next step for agents in vision-and-language navigation (VLN). This work tackles the long-standing generalization problem in VLN, aiming to enable agents to navigate unseen environments and bridge the simulation-to-reality gap. Additionally, his exploration of trajectory planning for robotic arms, using MQTT and PID algorithms, demonstrates a hands-on approach to control systems. Through these diverse projects, Wang is establishing himself as a versatile engineer and scientist, building intelligent systems that see, understand, and act.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3