CS3 Monthly Research Exchange

At each CS3 Monthly Research Exchange, three faculty and student researchers in the field of smart urban planning will take the stage to share their latest findings, breakthroughs, and urban projects.

Sharon Di, Associate Professor at Columbia University

• Presentation Abstract: Transportation digital twins have become increasingly popular tools to improve traffic efficiency and safety. However, the majority of effort nowadays is focused on the “eyes” of the digital twin, which is object detection using computer vision. I believe the key to empowering the intelligence of a transportation digital twin lies in its “brain,” namely, how to utilize the information extracted from various sensors to infer traffic dynamics evolution and devise optimal control and management strategies with real-time feedback to guide the transportation ecosystem toward a social optimum. My research aims to employ tools including machine learning and game theory to develop an urban transportation digital twin.

Abigail Joseph, Undergraduate Student at Florida Atlantic University

• Presentation Abstract: Integrating Unity with real data, AI technology, and C# scripting is used to a create digital twin environment of West Palm Beach, where multiple agents’ traveling behaviors are defined at a streetscape level, including walking at a particular speed, stopping at an intersection to look for cars before crossing, and avoiding other pedestrians. It is anticipated that the methods which produced these results will be used to simulate the interactions between vehicles and pedestrians with substantial accuracy.

Carl Vondrick, Associate Professor at Columbia University

• Presentation Abstract: Computer vision algorithms need to combine many skills — spatial, physical, mathematical, geometrical, and cognitive — in order to accurately analyze the visual world. In this talk, I will show how large generative models equip neural networks with these skills, thereby providing versatile representations for reconstructing 3D, answering questions, and recognizing objects. Through a series of experimental results, I will moreover show how this approach naturally provides inherent explainability of the decision making process, while also achieving state-of-the-art zero-shot performance across different tasks and benchmarks. In some cases, this framework can even perform super-human perception.

The CS3 Monthly Research Exchanges are internal and open only to CS3 affiliated students, faculty, and staff. If you are interested in learning more about the research happening at CS3, please contact our team at streetscapes@columbia.edu.