Cindy Wu

Website: https://xindiwu.github.io/

I am a second year master student at CMU’s Master of Science in Computer Vision program. During my undergrad, I was in Computer Science major (Special program for gifted young program) in Xi’an Jiaotong University. My research interest is in Multimodal machine learning, Trustworthy/Robust machine learning and their application in autonomous driving / robots. I had research internship in Megvii(Face++) and Snap Inc. before.

Project responsibilities: Defined the new task and benchmark for map maintenance, evaluating both fidelity and generalization. Leveraged the Argoverse dataset to define a comprehensive and rigorous set of evaluation criteria.
Proposed a contrastive cross-modal approach to dynamic street map construction from camera data. Trained the graph encoder and image encoder with a shared latent space building on recent advances in multimodal representation learning.

Kwun Fung(Raymond) Lau

Linkedin: https://www.linkedin.com/in/kflau/

I am a second year student at CMU’s Master of Computer Vision program. I received my bachelor’s degree from The Hong Kong Polytechnic University. Before joining the master program, I worked as an algorithm design engineer in the Hong Kong Applied Science and Technology Research Institute for 3 years.

My previous works mainly focus on applied research (i.e. applying computer vision algorithms to solve a use case and to optimize them on a Neural Processing Unit). I am comfortable in programming using C++ (11/14) and Python.

Project responsibilities: Survey sequential based approaches for graph generations(including GraphRNN, molecularRNN, HDMapGen etc), explore the limitations of the current existing sequential based approaches, metrics implementation, dataset/results visualizations and miscellaneous tasks.

Professor Deva Ramanan 

Academic supervisor of this capstone project. Deva is a CMU RI Professor and is a principle scientist at Argo AI and the director of the CMU Argo AI Center for Autonomous Vehicle Research.

Dr. Francesco Ferroni

Industrial supervisor of this capstone project. Francesco is a Software Engineering Manager II of Detection team at Argo AI.