Naji Khosravan
Computer Science Ph.D.
Senior Applied Science Manager (AI/ML)
ABOUT ME
I'm passionate about developing cutting-edge Computer Vision and Machine Learning technologies, turning research into real-world impact.
Currently, I lead the AI Media Insights team within Zillow Group’s AI Org, where we design and deploy advanced CV/ML models to enhance Zillow’s products.
I earned my Ph.D. in Computer Science from the Center for Research in Computer Vision (CRCV) at the University of Central Florida (UCF), specializing in Computer Vision and Deep Learning. My research focused on Human-Centered AI for analyzing volumetric medical scans using smart glasses. During my Ph.D. I had a couple of internships at Netflix Research working on Attention based methods for video and an internship at Sighthound, a startup company where I worked on representation learning for multi-object tracking.
My research interests lie at the intersection of Multimodal Learning, 3D Computer Vision, and Generative Models. I focus on advancing general-purpose visual intelligence through multimodal and multitask learning—enabling AI systems to develop a deeper understanding of the physical world, seamlessly integrate with active agents, and adapt to local data using privacy-preserving techniques like Federated Learning.
NEWS & HIGHLIGHTS
July 2024:
Giving an invited talk at Gradients Spaces at Stanford University.
June 2024:
We released ZInD-Tell enhancing Zillow Indoor Dataset with natural language descriptions of layout at CVPR'24.
June 2023:
I'm organizing CIVILS workshop in conjunction with CVPR'23.
June 2023
I'm co-organizing FedVision workshop in conjunction with CVPR'23.
June 2023:
Our paper Graph-Covis received the best paper award from OmniCV workshop at CVPR'23.
June 2022:
Our paper LASER on visual localization in textureless indoor scenes is selected as an Oral presentation at CVPR'22.
June 2022:
We lunched Zillow's AS GitHub landing page to open-source our research Projects.
June 2021:
We released (academic use only) Zillow Indoor Dataset (ZInD) the largest real dataset of 360 panoramas and floor-plans in CVPR'21.
SELECT PROJECTS
Tonmoay Deb, Lichen Wang, Zachary Bessinger, Naji Khosravan, Eric Penner, Sing Bing Kang,
CVPRW, 2024
Taotao Jing, Lichen Wang, Naji Khosravan, Zhiqiang Wan, Zachary Bessinger, Zhengming Ding, Sing Bing Kang
WACV, 2024
Negar Nejatishahidin, Will Hutchcroft, Manjunath Narayana, Ivaylo Boyadzhiev, Yuguang Li, Naji Khosravan, Jana Košecká, Sing Bing Kang
CVPRW, 2023, (OmniCV Best Paper Award)
Yu Yin, Will Hutchcroft, Naji Khosravan, Ivaylo Boyadzhiev, Yun Fu, Sing Bing Kang
ICMR, 2022
Zhixiang Min, Naji Khosravan, Zachary Bessinger, Manjunath Narayana, Sing Bing Kang, Enrique Dunn, Ivaylo Boyadzhiev
CVPR, 2022 (Oral Presentation)
Steve Cruz, Will Hutchcroft, Yuguang Li, Naji Khosravan, Ivaylo Boyadzhiev, Sing Bing Kang
CVPR, 2021
Rodney LaLonde, Naji Khosravan, Ulas Bagci
Advanced Intelligent Systems, 2021
Naji Khosravan, Shervin Ardeshir, Rohit Puri
CVPRW, 2019
Naji Khosravan
Ph.D. Dissertation, 2019
Naji Khosravan, Aliasghar Mortazi, Michael Wallace, Ulas Bagci
MICCAI, 2019
Naji Khosravan, Haydar Celik, Baris Turkbey, Elizabeth C Jones, Bradford Wood, Ulas Bagci
Medical image analysis journal, 2019
Naji Khosravan, Ulas Bagci
IEEE EMBC, 2018
Naji Khosravan, Haydar Celik, Baris Turkbey, Ruida Cheng, Evan McCreedy, Matthew McAuliffe, Sandra Bednarova, Elizabeth Jones, Xinjian Chen, Peter Choyke, Bradford Wood, Ulas Bagci
MICCAIW, 2017