Furqan Muhammad Khan



Activity recognition in cluttered environment

The objective was to recognize six different human activities performed in variety of settings and recorded from different viewing angles and to design an activity recognition system that gives near real-time speed for activity recognition in these settings. Space-time interest points were used in a graphical framework to achieve scale and viewpoint invariance while obtaining continuous action recognition in cluttered environment. The work was done at University of Southern California and published in WMVC 2009. (Publication information here)

Real-time tracking of air-borne targets

This work was done at Space and Upper Atmosphere Research Commission of Pakistan. The objective was to design a tracking system mounted on a moving platform that can track a variety of air-borne objects at a variety of speeds from airplanes to birds. System had to be fast and responsive and needed to be implemented on dedicated hardware. A paper titled, "Implementation of Real-Time Visual Tracking System," was published in proceedings of VISAPP 2008. (Publication information here)



Disclaimer: IRIS group, Computer Science Department, USC Viterbi School of Engineering or University of Southern California do not screen or control the content on this website and thus does not guarantee the accuracy, integrity, or quality of such content. All content on this personal website is provided by and is the sole responsibility of the person from which such content originated..