Back to Technology  |  More Profiles
Watch faculty video
Explore More: Read Online Brochure

Amit K. Roy-Chowdhury

Professor of Electrical Engineering
Amit K. Roy-Chowdhury
Creating Better Video Analysis Technology
Looking at any given image or video sequence, Amit Roy-Chowdhury is looking to understand the content. Similar to the brain’s processing and interpreting of images, his research seeks to develop automated algorithms that not only capture and store images, but understand the content of those images.

Prof. Roy-Chowdhury's research has been supported by various agencies including the National Science Foundation, Office of Naval Research, Army Research Office, DARPA, National Endowment for the Humanities, and private industries like CISCO and Lockheed-Martin. His recent book on Camera Networks provides an overview of current research in the field.

Areas of Expertise

Areas of Expertise:
  • Computer Vision and Image Processing
  • Statistical Pattern Recognition
  • Machine Learning
  • Multimedia
  • Imaging/Non-imaging Sensor Networks
  • Biological Image Processing
College: Department: Press Release/Article: Profile:

Select Honors and Distinctions

  • Best Student Paper Award (with his graduate student), ICASSP 2006.
  • Associate Editor, IEEE Trans. on Cybernetics


Q: How might your research change the way surveillance footage is viewed?
We are developing methods for tracking objects in video and recognizing their activities. One of our success areas is wide-area camera networks such as those found in an airport or shopping mall, where data from many cameras is stored. In these cases, most analysis is done after the fact and only if there is a need to review footage, whereas we’re working to embed intelligence into the system so that security personnel are alerted to pay attention to specific screens in real time.

Q: What are the real-life implications of your research?
Results we’ve obtained from our research can be used in different applications. In regards to security surveillance, our results can be used to analyze data and trends to model various kinds of human behaviors. As well, biologists can track cells and growing tissues and in the domain of art, historians are able to recognize faces in portraits from many centuries ago.

Q: What are the benefits of using image analysis on social media sites?
Analysis of images, along with the text, on social media sites like Twitter can provide real-time information about events like music concerts or long-term trends like changing traffic patterns. These could be used for various purposes, from more effective law enforcement to better planning.

Q: Are there any privacy issues or concerns with this technology?
The use of computational tools for the analysis of data, video or otherwise, do entail natural and valid concerns about privacy. Technology can be used to address some of these. Overall, it requires a combination of scientists, engineers and policy makers coming together to devise effective rules and regulations.

Q: What interests have driven your research?
Overall in my research I’m interested in two directions: the fundamental science behind the analysis of the content of images and videos, and to look at specific application domains where this research can be effectively implemented, such as biomedical purposes to help better diagnose diseases or understanding basic biological processes, social media, environmental concerns such as capturing videos in remote places to document animal behavior, transportation with cameras in cars monitoring driver behavior or on highways monitoring traffic patterns.

There are significant possibilities for our research to be used in many applications. With funding from the National Science Foundation and the Office of Naval Research, and in collaboration with many of my colleagues, we have been working on scene understanding using a distributed network of cameras spread over a wide area. We have obtained very promising theoretical results and shown their implementation in UCR’s camera network facility. Such methods could one day be used for applications like disaster recovery, whereby a group of robots equipped with cameras could work alongside first responders and access areas that may be too dangerous for humans.

Q: What is the future of video analysis technology?
In regards to security and surveillance, video analysis technology will automate a lot of processes. Automated analysis will help security officers better focus their attention to specific, necessary areas. Commercial multimedia applications can use video content analysis in order to feed recommended videos to users depending on interests. As well, researchers in biology and medicine can leverage upon the analysis of images to diagnose specific conditions or obtain a more precise and quantitative understanding of biological phenomena.

Q: What does Living the Promise mean to you?
Living the Promise means working on solutions to problems that will have a positive impact on society. The knowledge learned in the pursuit of this research is transferred to the next generation so that the promise really lives on.

Q: Why is UC Riverside a great place to do research?
UCR is a great place to do research because the overall environment here is helpful to research and encourages me to explore my own ideas. I’m lucky to have excellent students to work with and very supportive staff, who have made my research possible.

Q: What movies, documentaries or TV programs have you recently found interesting and/or inspiring?
PBS and National Geographic documentaries.

Amit K. Roy-Chowdhury “Video analysis could lead to significant benefits in a number of applications. It could help law enforcement analyze security footage more efficiently, organize and search through huge video repositories, or help in better medical diagnoses.”

—Amit K. Roy-Chowdhury
Living The Promise Report Explore More: Read Online Brochure