Boyang Deng

Papers

Total Citations

H-Index

About

Boyang Deng is a researcher at the forefront of 3D computer vision and robotics, with a focus on scalable world modeling for autonomous systems. His primary research areas include 3D object detection, implicit neural representations, and generative models for simulation. Deng’s major contribution is the development of GINA-3D, a pioneering framework that learns to generate implicit neural assets directly from real-world sensor data, enabling the automatic creation of realistic 3D environments for robotic testing and validation. This work addresses the critical bottleneck of manual environment creation in autonomous driving research, with his most-cited paper on this topic accumulating 16 citations. Additionally, Deng has made significant strides in rethinking 3D object detection from an egocentric perspective, emphasizing how detections impact an agent’s behavior and safety rather than just raw accuracy. His 2021 paper on this topic, with 9 citations, provides a novel framework for evaluating detection performance in safety-critical applications. Deng’s work bridges the gap between perception and simulation, offering scalable solutions that are essential for advancing autonomous driving and robotics. His research is widely recognized for its practical impact on developing robust, real-world-ready systems.