Ross Greer
Google Scholar | MI3 Lab
I am an Assistant Professor in the Computer Science & Engineering Department at the University of California Merced.
I use computer vision & artificial intelligence to create systems that help keep people safe and create or enhance human-interactive capabilities. My research is most often applied to the domain of Safe Autonomous Driving. Broadly speaking, my research focuses on improving computational intelligence by defining new information representations, architectures, and metrics for machine learning. The goal of my research is to build systems which are adaptable to the open world, robust to the long-tail problem and curse of rarity, and safe around chaotic human agents. Methods I develop are used to quantify uncertainty, importance, and novelty in learning systems, improving data utility and curation.
In autonomous driving, my research is applied to form robotic plans and decisions for autonomous agents, understand driver state through analysis of visible and multimodal features (e.g. gaze and hand activity), learn salience of (and relationships between) road objects and drivers, and predict vehicle trajectories in real-world driving scenarios.
Additional areas of my research facilitate human-human and human-machine musical interactions, using AI to quantify and learn not only performer behaviors and symbolic & acoustic musical patterns, but also latent representations of "imagination", "creativity", and expression.
My Ph.D. was supervised by Mohan Trivedi in the Laboratory for Intelligent & Safe Automobiles (LISA) at the University of California San Diego, following the MS degree at UCSD and BS and BA degrees at UC Berkeley. I conduct further research with Shlomo Dubnov through the Center for Research in Entertainment & Learning (CREL).
My research is supported by the Qualcomm Innovation Fellowship, IRCAM's Project REACH, as well as the generous support of lab sponsors Toyota CSRC, AWS, and the UCOP ILTI grant.
Outside of my research, I advise and conduct the orchestra of the Symphonic Student Association, and field direct & write music for the University of California Marching Band.
Publications
Preprints of many of my papers are available at cvrr.ucsd.edu.
News & Media
Three of our papers were accepted for presentation at the CVPR workshop on Vision-Language for Autonomous Driving and Robotics.
Two of our posters have been accepted for presentation at the TRB Automated Road Transporation Symposium.
I am a 2024 recipient of the Interdisciplinary Research Award, awarded by the UC San Diego GPSA.
I am the 2024 recipient of the Henry Booker Award for Exemplary Ethical Engineering, awarded by the Jacobs School of Engineering.
I am the recipient of a Postdoctoral Networking Fellowship from the Germany's Academic Exchange Service; my research bio is featured in their program.
Professor Shlomo Dubnov presented our research in AI & Visual Cues for Musical Improvisation at the 10th annual Improtech conference.
My colleague Lulua Rakla and I won the Grand Prize for the AWS Automotive Day competition at IEEE Intelligent Vehicles Symposium, for implementing a drowsy driving detection system using AWS cloud solutions.
My poster for Safe, Multi-View Activity Recognition by Tracking Hands: Robust Visual Learning with Missing Data won the Grand Prize and Department Best Poster Award at the 2023 Jacobs Research Expo. Thank you to collaborators Lulua Rakla, Akshay Gopalkrishnan, Anish Gopalan, and Professor Trivedi. Press Release
Our research project SMART Hands was selected by NHTSA as the North American representative for the final round of the Safety Design Competition for the 2023 International Technical Conference on the Enhanced Safety of Vehicles in Yokohama, Japan.
Two of our research projects (CHAMP, SMART Hands) were selected by NHTSA as regional finalists in the Safety Design Competition for the 2023 International Technical Conference on the Enhanced Safety of Vehicles.
My Polygence blog post High School Computer Science Research: The Complete Guide from “Hello, World!” to the Real World has climbed the ranks of Google Search!
My research partner Kuan-Lin Chen & I are finalists for winners of (!) the 2022 Qualcomm Innovation Fellowship. Our proposal is on Active Learning for Autonomous Driving Datasets.
My poster for Trajectory Prediction with a Novel Lane-Heading Auxiliary Loss won a Best Poster Award at the 2021 Jacobs Research Expo. Thank you Nachiket & Professor Trivedi for your guidance! Press Release
Our collaboration with Amazon's Machine Learning Solutions Lab was featured in an AWS blog post.
My work with Professor Shlomo Dubnov in restoring nonverbal communication to virtual classrooms was featured in NewScientist magazine, Engineering.com, and CSE News,
I was interviewed by Pianist Tiffany Poon for an episode of Together With Classical, where we discussed the intersection of A.I. & Music.
I spoke at the Winter 2021 Educational Innovation Expo @ UCSD.
Teaching
I will begin teaching courses at UC Merced soon, and will update this page accordingly with links.
In the meantime, students of UCSD ECE 172A, 180, 253, 285, or CSE 190 are welcome to reach out regarding letters of support or recommendation related to my teaching duties in these courses.
For those who enjoyed CSE 190, or my past lectures on neural networks, please check out the textbook I wrote with Shlomo Dubnov - Deep and Shallow: Machine Learning in Music and Audio.