Embedded Files
CVPR 2026 Workshop Precognition
Chengbo Zang*, Wangshu Zhu*, Mehmet Kerem Turkcan, Siyi Liu, Hanlin Wang, Xudong Chen, Boshra Khalili, Noshin Saiyara Ahmad, Peter Jing Jin, Javad Ghaderi, Gil Zussman, Zoran Kostic
A data synthesis pipeline for generating realistic traffic scenes and safety-critical rare events under natural language instructions while providing agent relation annotations.
NeurIPS 2025 Workshop GenAI4Health
Yu-Wen Chen, William Ho, Sasha M Vergez, Grace Flaherty, Pallavi Gupta, Zhihong Zhang, Maryam Zolnoori, Margaret V. McDonald, Maxim Topaz, Zoran Kostic, Julia Hirschberg
LLMs for automated health assessment and ALMs for extracting vocal biomarkers in home healthcare settings.
Chengbo Zang, Mehmet Kerem Turkcan, Gil Zussman, Zoran Kostic, Javad Ghaderi
A general framework for adaptively collecting the training data to achieve robust model performance across multiple data distributions, with theoretical guarantees and experimental verifications on multiple tasks.
Yuyang Li, Devika Gumaste, Mehmet Kerem Turkcan, Javad Ghaderi, Gil Zussman, Zoran Kostic
A distributed architecture for VLMs that addresses the limitations of edge deployment by partitioning model components between edge devices
and central servers.
William Ho, Mehmet Kerem Turkcan, Gil Zussman, Javad Ghaderi, Zoran Kostic
A SAM-based open-vocabulary classifier for zero-shot image tagging for an urban traffic intersection.
Chengbo Zang, Mehmet Kerem Turkcan, Gil Zussman, Javad Ghaderi, Zoran Kostic
A data-driven traffic simulation of a metropolitan street intersection leveraging statistical priors and deep-learning-based trajectory forecasting models.
CVPR 2024 Demo
Mehmet Kerem Turkcan, Mahshid Ghasemi Dehkordi, Sofia Kleisarchaki, Thomas Calmant, Levent Gürgen, Javad Ghaderi, Gil Zussman, Zoran Kostic
A comprehensive testbed for the application of computer vision in New York City street intersections.
CVPR 2024 Embedded Vision Workshop
Zoran Kostic
Understanding "real-time" for urban streetscape video-based applications.
PEER-REVIEWED PUBLICATIONS
PEER-REVIEWED PUBLICATIONS
OTHER PUBLICATIONS, TALKS AND PRESENTATIONS
OTHER PUBLICATIONS, TALKS AND PRESENTATIONS
X. Di et al., “AI-Powered Urban Transportation Digital Twin: Methods and Applications,” Dec. 30, 2024, arXiv: arXiv:2501.10396. doi: 10.48550/arXiv.2501.10396.
M. K. Turkcan, Y. Li, C. Zang, J. Ghaderi, G. Zussman, and Z. Kostic, “Boundless: Generating Photorealistic Synthetic Data for Object Detection in Urban Streetscapes,” Sep. 26, 2024, arXiv: arXiv:2409.03022. doi: 10.48550/arXiv.2409.03022.
N. S. Pargoo et al., “The Streetscape Application Services Stack (SASS): Towards a Distributed Sensing Architecture for Urban Applications,” Jan. 13, 2025, arXiv: arXiv:2411.19714. doi: 10.48550/arXiv.2411.19714.
M. K. Turkcan, M. Ballo, F. Filicori, and Z. Kostic, “Towards Suturing World Models: Learning Predictive Models for Robotic Surgical Tasks,” Mar. 16, 2025, arXiv: arXiv:2503.12531. doi: 10.48550/arXiv.2503.12531.
M. Ghasemi, Z. Yang, M. Sun, H. Ye, Z. Xiong, J. Ghaderi, Z. Kostic, and G. Zussman, “Demo: Video-based social distancing evaluation in the COSMOS testbed pilot site,” in Proc. ACM MOBICOM’21, 2021.
Z. Yang, M. Sun, H. Ye, Z. Xiong, G. Zussman, and Z. Kostic, “Birds eye view social distancing analysis system,” arXiv:2112.07159 [cs.CV], Dec. 2021.
Richard Samoilenko, Ashvin Jagadeesan, Zihao Xiong, Hongzhe Ye, Zhengye Yang, Zoran Kostic, "Optimizing YOLOv4 Object Detection in Smart-City Traffic Intersections," at Columbia University Data Science Day, Wednesday, April 21, 2021.
Mahshid Ghasemi, Zhengye Yang, Mingfei Sun, Hongzhe Ye, Zihao Xiong, Zoran Kostic, Javad Ghaderi, Gil Zussman, “COSMOS Testbed – Proximity Detection and Social Distancing Estimation in COVID-19 Pandemic,” Workshop on Challenges for Digital Proximity Detection in Pandemics: Privacy, Accuracy, and Impact, NIST, January 26, 2021 - January 28, 2021.
Zhengye Yang, Mingfei Sun, Hongzhe Ye, Zihao Xiong, Gil Zussman, Zoran Kostic, "Privacy Preserving Social Distancing Analysis in a Metropolis," in Machine Learning in Science & Engineering, Data Science Institute at Columbia University, Virtual Conference, December 14 - 15, 2020.
Zoran Kostic, “Object Detection in Smart City Intersections: Real-time Considerations,” 2nd COSMOS Research Community Workshop, Oct 14-15, 2020.
Zoran Kostic, “COSMOS Smart Traffic Intersection: High-Bandwidth Low-Latency Applications using AI and Edge Computing Infrastructure - A Snapshot on Real Time Consideration,” IEEE 5G World Forum - AI/ML Track, Sept 10-12, 2020.
COSMOS Smart Traffic Intersection: High-Bandwidth Low-Latency Applications using AI and Edge Computing Infrastructure - A Snapshot on Real Time Consideration, IEEE 5G World Forum - AI/ML Track, Sept 10-12, 2020.
Zhengye Yang, Emily Bailey, Shiyun Yang, Gil Zussman, Ivan Šeškar, Zoran Kostic, "Smart City: Deep Learning in the Edge-Cloud for COSMOS Smart Intersection," Data Science Day, Columbia University, Tuesday, March 31, 2020.
Zoran Kostic, “Deep Learning in the Edge-Cloud for COSMOS Smart Intersections,” IEEE Workshop on AI/ML in Networks and Cloud; Feb 17, 2020; Rutgers University Inn and Conference Center; New Brunswick, New Jersey.
COSMOS team, “COSMOS REM/RET program and educational toolkit,” nominated for the Global Mobile Awards - category on YoMo STEAM Activity for Young People, at the 2020 Mobile World Congress (event postponed).
Z. Kostic, “Smart City Intersections,” COSMOS Experimenters Workshop, Rutgers University, May 21, 2019.
Z. Kostic, COSMOS PIs, "Panel: The COSMOS Advanced Wireless Testbed: integrating Gigabit wireless access with low latency edge computing," EDGE 10Conference on Edge & Fog, Friend Center, Princeton University, NJ, May 17, 2019, http://edge10.princeton.edu.
Zoran Kostic, "COSMOS Project - Opportunities for Health Applications," Columbia Data Sciences Institute (DSI) Health Analytics workshop, Columbia University, New York City, March 1, 2019.
Zoran Kostic and Ivan Seskar, "Smart City Intersections in Megapolises," Sense, Collect and Move Data Seminar, Columbia Data Sciences Institute (DSI), New York City, Oct. 12, 2019.
Zoran Kostic et. al, "Pedestrian and vehicle tracking in bird’s eye view videos," Smart Cities Center Poster Session, Columbia Data Science Institute, New York City, November 14th, 2018 .
Tianyao Hua and Zoran Kostic, "Pedestrian and vehicle tracking in bird’s eye view videos," Columbia Data Sciences Institute (DSI) Summer Student 2018 EE Expo, New York City, 2018.
Zoran Kostic, "COSMOS overview," Columbia Data Sciences Institute (DSI) retreat, Tarrytown, New York, October 7-8, 2018.
Zoran Kostic, “Smart City: Deep Learning Models for Car-Pedestrian Interaction,” NVIDIA GPU Technology Conference, San Jose, March 25-29, 2018.
Zoran Kostic, "Internet of Things - Physical Data Analytics," Workshop Chair, IEEE International Conference on Sensing, Communication and Networking, 27-30 JUNE 2016, LONDON, UK.
Zoran Kostic, "The Next Big Revolution: Internet of Things," Invited talk and panel moderator, Japan Society, September 26, 2016, NYC.
Google Sites
Report abuse