Mobile Applications Lab - University of Toronto

Welcome to UofT's Mobile Applications Lab (APL). APL research focuses on novel algorithms, technologies, and systems that use the advanced sensing and visual+acoustic+tactile interaction mechanisms of modern smartphones/tablets. In other words, we explore ways to create disruptive mobile apps with a particular focus on sensing and information processing.

Our physical lab, in the Bahen Center for Information Technology (Room 4158), hosts a variety of dedicated stations for application exploration, simulation, and testing, including iOS and Android equipped development stations (more platforms to come soon). This facility is open to any and all UofT student/employee/faculty who are interested in mobile applications research (if you are interested to use our facility, please send an email to parham@ecf.utoronto.ca).

If you are interested in creating and learning about mobile apps, please visit the ECE1778 course website.

Research Spotlight

Here are a few recent projects from our lab:

Real-Time Object Tracking on iPhone
Amin Heidari, Parham Aarabi: Real-Time Object Tracking on iPhone. ISVC (1) 2011: 768-777

A novel real-time object tracking algorithm is proposed which tracks objects in real-time on an iPhone platform. The system utilizes information such as image intensity, color, edges, and texture for matching different candidate tracks. The tracking system adapts to changes in target appearance and size (including resizing candidate tracks to a universal depth-independent size) while running at 10-15FPS tracking rate. Several experiments conducted on actual video are used to illustrate the proposed approach.

Tiny Videos: A Large Data Set for Nonparametric Video Retrieval and Frame Classification
Alexandre Karpenko, Parham Aarabi: Tiny Videos: A Large Data Set for Nonparametric Video Retrieval and Frame Classification. IEEE Trans. Pattern Anal. Mach. Intell. 33(3): 618-630 (2011)

In this paper, we present a large database of over 50,000 user-labeled videos collected from YouTube. We develop a compact representation called 'tiny videos' that achieves high video compression rates while retaining the overall visual appearance of the video as it varies over time. We show that frame sampling using affinity propagation - an exemplar-based clustering algorithm - achieves the best trade-off between compression and video recall. We use this large collection of user-labeled videos in conjunction with simple data mining techniques to perform related video retrieval, as well as classification of images and video frames. The classification results achieved by tiny videos are compared with the tiny images framework for a variety of recognition tasks. The tiny images data set consists of 80 million images collected from the Internet. These are the largest labeled research data sets of videos and images available to date. We show that tiny videos are better suited for classifying scenery and sports activities, while tiny images perform better at recognizing objects. Furthermore, we demonstrate that combining the tiny images and tiny videos data sets improves classification precision in a wider range of categories.

Intelligent Ad Resizing
Anthony P. Badali, Parham Aarabi, Ron D. Appel: Intelligent ad resizing. WWW 2010: 1053-1054

Currently, online advertisements are created for specific dimensions and must be laboriously modified by advertisers to support different aspect ratios. In addition, publishers are constrained to design web pages to accommodate this limited set of sizes.

As an alternative we present a framework for automatically generating visual banners at arbitrary sizes based on individual prototype ads. This technique can be used to create flexible visual ads that can be resized to accommodate various aspect ratios. In the proposed framework image and text data are stored separately. Resizing involves selecting a sub-region of the original image and updating text parameters (size and position). This problem is posed within an optimization framework that encourages solutions which maintain important structural properties of the original ad. The method can be applied to advertisements containing a wide variety of imagery and provides more flexibility than existing solutions. In addition to positive qualitative resizing results, the proposed framework outperformed alternative techniques by over 20% in a user study.