The DVMM Lab at Columbia University is dedicated to research of multimedia content understanding, computer vision, machine learning, and signal processing. Our primary goal is to develop intelligent systems that can harness rich information from the vast amount of visual data such as those shared on the open Web, collected through pervasive imaging devices, or stored in gigantic content archives. A consistent theme of our research is to turn unstructured multimedia data into useful and searchable information. Our work on content-based visual search in the early 90's, VisualSEEk and VideoQ, set the foundation of this vibrant area. Over the years, we continued to create new theories, algorithms and systems for image/video recognition, multimodal analysis, multimedia ontology construction, graph-based semi-supervised learning, and learning-based compact hashing for big data indexing. We also developed novel applications for multi-source news video aggregation, mobile visual search, and brain machine interfaces for large-scale information triage. With many high-impact publications and more than 30 issued patents, our results have been broadly distributed as open research resources and incorporated in practical systems in industry and government.

Please visit the research page and the publication list to learn more about our work in

multimedia search and retrieval
machine learning and computer vision
mobile communication and applications
media security and forensics
benchmarking and standard

DVMM news (All news)
01/2016 Prof. Chang received Honorary Doctorate from the University of Amsterdam (more)
10/2015 DVMM won Best Applied Paper Award in Semantic Web Conference 2015 (more)
09/2015 Published survey paper on compact hashing for large-scale retrieval (more)
09/2015 A new project on Multilingual Visual Sentiment Ontology launched (more)
09/2015 DVMM launched EventNet large-scale structured video event ontology and detectors (more)
04/2015 DVMM alum John R. Smith inducted IBM Fellow, the company's most prestigious technical honor (more)
01/2015 Prof. Chang gave distinguished lecture at UCLA (more)
09/2014 NewsRover project won the 2nd demo prize at NYC Media Lab Summit (more)
09/2014 DVMM Demos at NYC Media Lab Summit: News Rover, Assistive Image Comment Robot (more)
07/2014 DVMM alum Hari Sundaram joins UIUC CS as tenured faculty (more)
06/2014 DVMM alum Shahram Ebadollahi became Vice President, Health Informatics Research & Chief Science Officer, IBM Healthcare (more)
04/2014 Subh and Felix won Best Paper Award at ICMR 2014 (more)
03/2014 Felix Yu received 2014 IBM PhD Fellowship Award (more)

* This is the new version of DVMM website, for the previous version please go to

bar

For problems or questions regarding this web site contact The Web Master. Last updated: Jan., 2016.