Jiteng Mu

Shandong University, UC San Diego Health System

Papers

Total Citations

H-Index

About

Jiteng Mu is a researcher at the forefront of robotics and computer vision, with a focus on enabling robots to perceive and interact with their environments more intelligently. His work bridges the gap between real-time object recognition and generalizable scene understanding for autonomous systems. Mu’s early contributions include developing a novel, efficient algorithm for arbitrary colored ball recognition in humanoid soccer robots, addressing the critical challenges of computational complexity and false positive rates in dynamic RoboCup environments. This foundational work, cited 6 times, demonstrates his commitment to practical, real-world robotic perception. More recently, Mu has tackled the open problem of mobile manipulation with his 2025 paper on learning generalizable feature fields. This work proposes a unified representation for objects and scenes, allowing robots to seamlessly integrate navigation and manipulation capabilities—capturing intricate geometry while understanding fine-grained semantics. With 2 citations already, this emerging research promises to advance how robots operate in complex, unstructured spaces. Mu’s trajectory from efficient ball recognition to holistic scene representation marks him as a rising innovator in embodied AI, pushing the boundaries of what autonomous robots can achieve in dynamic environments.