Category Archives: Previous Funded Research Projects

Sign-Language Recognition from RGBD Data

PI: Dr. Mohamed Elsayed
Co-PI: Dr. Marwan Torky

Funding Agency: Microsoft ATLc
Duration: 12 months


Project Abstract

One of the most challenging problems in computer vision research is visual recognition and its related tasks, such as object classification, localization, activity, scene and event classification, etc. However, such challenging problems have benefited a lot from recent advances in sensing technologies, such as cheap RGBD sensors (e.g. Microsoft Kinect). The merit of using depth sensors is straight forward. While the original image capturing is a projection of the 3-D world into a 2-D image plane (which results in ambiguities), the RGBD data aims to reduce ambiguity by giving an easily-calibrated depth data to the captured pixels in the 2-D image.

Currently, many indoor applications such as 3-D reconstruction of indoor scenes, robot navigation, and activity recognition have started using Kinect-like sensory data. In this research project, we address the problems of action recognition (sign language, in particular) using RGBD data. The particular aims are the following: First, we collect a dataset for isolated-word sign language using a Kinect sensor. Second, we develop and test algorithms that apply machine learning on the collected dataset to recognize sign language from the user’s skeleton movement and hand and face shapes.

Personalized Microblogs Corpus Recommendation

PI: Dr. Hesham El Mongy
Funding Agency:Microsoft ATLC
Duration: 12 Months
Status: Started on January 2014

Project Abstract

Microblogs are special virtual social network web-based applications. Users of microblogs are allowed to post relatively short messages (corpuses) compared to regular blogs. This encouraged many users to become more active, as the effort they need to put to post a message is very small. On the other hand, following the microblogs is becoming more challenging as users can receive thousands of corpus updates every day. Going through all the corpuses updates is a time consuming process and affects the user’s productivity in real life, especially for the users who have a lot of followees and thousands of tweets arriving at their timelines every day. In this project, we propose a personalized recommendation system that will aim to give the user a summary of all received corpuses. Considering the fact that the user interests change over time (and location), this summary should be based on the user’s level of interest in the topic of the corpus at the time of reception. Our method considers three major elements: users’ dynamic level of interest in a topic, user’s social relationship such as the number of followers, their real geographical neighborhood, and other explicit features related to the publishers’ authority and the tweet’s content.