Real-time detection of commercials on television

For my undergraduate honours thesis I conducted research into the unbuffered real-time detection of commercials on television with a view to muting the volume when ads are being broadcast. The research itself dealt with examining the features required to enable robust real-time detection. I developed a sophisticated video analysis and processing framework to underpin experimentation and compilation of results.

The following screenshots show the system running live (click on one to see the full-res image):


I wrote a Windows-based video analysis and processing framework to underpin the research I undertook for my undergraduate thesis.

Some of the features it boasts:

Click on one of the following screenshots to see the full-res image:

The left-hand screen shows the main controller interface (middle) with the video preview above it. On the left side of the video preview is an example of a detector: the hard-cut detector. It outputs instantaneous information to the luminosity histogram on the right of the video preview,
and time-series information to the graph below the controller window. The right-hand screen shows the event viewer displaying four different detector (analysis) runs performed on portions of the input video. Here, each shows the before & after frame for a detected hard cut.

The controller window, video preview and event viewer can also be seen here. In addition, the DirectShow filter graph information window is visible below the controller, and on its right is the manual event input window. Overlaid upon the event viewer is the event list window, which displays events for each detector (analysis) run and enables a user to select the events/runs that should be shown, edited, etc.