Computer Science and Electrical Engineering
University of Maryland, Baltimore County
Ph.D. Dissertation Defense

A Rapidly Deployable Image Classification System Using Feature Views

Adrian Rosebrock

9:00am Friday, 18 April 2014, ITE 346, UMBC

Constructing an image classification system using strong, local invariant descriptors is both time consuming and tedious, requiring much experimentation and parameter tunings to obtain an adequate performing model. Furthermore, training a system in a given domain and then migrating the model to a separate domain will likely yield poor performance. As the recent Boston Marathon attacks demonstrated, large, unstructured image databases from traffic cameras, security systems, law enforcement officials, and citizens can be quickly amassed for authorities to review; however, reviewing each and every image is an expensive undertaking, in terms of both time and human effort. Inherently, reviewing crime scene images is a classification task. For example, authorities may want to know if a given image contains a suspect, a suspicious package, or if there are injured people in the photo. Given an emergency situation, these classifications will be needed as quickly and accurately as possible. In this work we present a rapidly deployable image classification system using “feature views”, where each view consists of a set of weak, global features. These weak global descriptors are computationally simple to extract, intuitive to understand, and require substantially less parameter tuning than their local invariant counterparts. We demonstrate that by combining weak features with ensemble methods we are able to outperform current state-of-the-art methods or achieve comparable accuracy with much less effort and domain knowledge. We then provide both theoretical and empirical justifications for our ensemble framework that can be used to construct rapidly deployable image classification systems called “Ecosembles”.

Finally, we recognize the fact that image datasets give us the relatively unique opportunity to extract multiple feature representations through the use of various descriptors. In situations where the original dataset is not available for further feature extraction or in cases where multiple feature views are ambiguous (such as predicting income based on geographical location and census data) the Ecosemble method cannot be applied. In order to extend Ecosembles to arbitrary datasets of diverse modalities, we introduce artificial feature views using kernel approximations. These artificial feature views are constructed from a single representation of the data, alleviating the need to explicitly extract multiple feature views. We then apply artificial feature views to a diverse range of non-image classification datasets to demonstrate our method is applicable to multiple modalities, while still outperforming current state-of-the-art methods.

Committee: Drs. Tim Oates (chair), Jesus Caban, Tim Finin, Charles Nicholas, Jian Chen

This March, while most college students were enjoying freedom from the classroom or even sunny days on the beach, a number of UMBC students instead chose to spend their week off from school in service learning experiences through Alternative Spring Break (ASB).

These week-long service trips introduce students to the broad range of challenges our society faces and teach them how they can help change things for the better. Thanks to donors like you, our students were volunteering and serving for projects and organizations like: 

• The Gesundheit Institute
•  Through the Eyes of a Refugee
• The ARC of Carroll County
• Will Hope for Change
• A Better Impact

Your gifts ensured that students could take advantage of these life-changing education opportunities, so thank you! They truly will carry these memories and lessons throughout their lives.

“My ASB experience was truly memorable. I loved seeing that small actions can make an impact and I had the best time working with other students in the program,” says student Erin Wells who went on the A Better Impact trip. “I have grown so much from just one week, thank you for this opportunity.”

See more photos from the Alternative Spring Break trips!

Every week we feature a set of comics created exclusively for WDD.

The content revolves around web design, blogging and funny situations that we encounter in our daily lives as designers.

These great cartoons are created by Jerry King, an award-winning cartoonist who’s one of the most published, prolific and versatile cartoonists in the world today.

So for a few moments, take a break from your daily routine, have a laugh and enjoy these funny cartoons.

Feel free to leave your comments and suggestions below as well as any related stories of your own…

Customer or suspect

The walking print dead

Compulsory financial planning

Can you relate to these situations? Please share your funny stories and comments below…

Smarty Bundle: Crests Logo Templates and Badges – only $15!

This March, while most college students were enjoying freedom from the classroom or even sunny days on the beach, a number of UMBC students instead chose to spend their week off from school in service learning experiences through Alternative Spring Break (ASB).

These week-long service trips introduce students to the broad range of challenges our society faces and teach them how they can help change things for the better. Thanks to donors like you, our students were volunteering and serving for projects and organizations like: 

• The Gesundheit Institute
•  Through the Eyes of a Refugee
• The ARC of Carroll County
• Will Hope for Change
• A Better Impact

Your gifts ensured that students could take advantage of these life-changing education opportunities, so thank you! They truly will carry these memories and lessons throughout their lives.

“My ASB experience was truly memorable. I loved seeing that small actions can make an impact and I had the best time working with other students in the program,” says student Erin Wells who went on the A Better Impact trip. “I have grown so much from just one week, thank you for this opportunity.”

See more photos from the Alternative Spring Break trips!

Justin Jacobs, a doctoral candidate in statistics at UMBC, has won the prestigious Presidential Early Career Award in Science and Engineering (PECASE). This is the highest honor bestowed by the U.S. government on science and engineering professionals in the early stages of their independent research careers, and recognizes Justin’s work with the intelligence community.

Justin received the award from Director of National Intelligence James Clapper on January 23, 2014, and he will be recognized by President Obama in a ceremony at the White House later this month.

At UMBC, Justin is being co-advised by Anindya Roy and John Zweck (now at UT Dallas), and plans to graduate this May. His dissertation is titled, “Density Estimation on Differential Manifolds.” He is completing his Ph.D. while employed with the National Security Agency.

President Clinton established the PECASE award program in 1996. The White House website notes, “Awardees are selected for their pursuit of innovative research at the frontiers of science and technology and their commitment to community service as demonstrated through scientific leadership, public education, or community outreach.” Learn more about PECASE and this year’s recipients in this White House Press release.

Congratulations, Justin, for this is a remarkable achievement!

Image: Director of National Intelligence James Clapper (far right) presents UMBC’s Justin Jacobs (center right) with his PECASE plaque. Photo courtesy of Justin Jacobs.