Parth Agrawal
I’m a Software Engineer at Google AR, where I develop the perception and motion tracking systems that power immersive augmented reality experiences for millions of users.
Previously at the UCLA Vision Lab, I explored the fascinating world of 3D vision and adversarial perturbations under the guidance of Dr. Stefano Soatto and Dr. Alex Wong. Additionally, I was a member of Dr. Ankur Mehta’s LEMUR lab, where I contributed to the development of autonomous blimps.
My research interests lie at the intersection of computer vision and machine learning, with a specific focus on 3D vision and sensor fusion. My work is motivated by a desire to understand human perception and develop intelligent systems that enhance human capabilities. My central focus is to create perception systems that can learn continuously and accurately interpret the 3D scene through geometry, enabling machines to perceive the world in a manner akin to human perception.
When I’m not busy moving matrices and manipulating images, you can find me behind a drum set, exploring new hiking trails, or playing table tennis. I’m always up for trying new activities and expanding my horizons.
Publications
Tong Ke, Parth Agrawal, Yun Zhang, Weikun Zhen, Chao X Guo, Toby Sharp, Ryan C Dutoit. “PC-SRIF: Preconditioned Cholesky-based Square Root Information Filter for Vision-aided Inertial Navigation”. Submitted to International Conference on Robotics & Automation 2025.
Tian Yu Liu*, Parth Agrawal*, Allison Chen*, Byung-Woo Hong, and Alex Wong. “Monitored distillation for positive congruent depth completion”. In European Conference on Computer Vision, pp. 35-53. Springer, Cham, 2022. (* Equal Contribution)
Zachary Berger*, Parth Agrawal*, Tian Yu Liu, Stefano Soatto, and Alex Wong. “Stereoscopic Universal Perturbations across Different Architectures and Datasets”. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 15180-15190. 2022. (* Equal Contribution)
Zhaoliang Zheng, Jiahao Li, Parth Agrawal, Zhao Lei, Aaron John-Sabu, and Ankur Mehta. User Based Design and Evaluation Pipeline for Indoor Airships. 39th IEEE Conference on Robotics and Automation (ICRA) 2022. Under Review.
Services
- AAAI, ICPR
Awards and Honors
Member, UPE
- Member of Upsilon Pi Epsilon, UCLA Computer Science honor society. Top 1/3rd of the class.
Member, HKN
- Member of Eta Kappa Nu, UCLA Electrical Engineering honor society. Top 1/4th of the class.
Member, TBP
- Member of Tau Beta Pi, UCLA Engineering honor society. Top 1/8th of the class.