A method of building an interactive display related to TV programs in an internet protocol television (IPTV) environment includes parsing a plurality of TV programs provided in an IPTV environment for display to a user. The step of parsing includes capturing an image in each TV program, identifying video content in each captured image, interpreting the video content, capturing audio data, identifying an audio content in each captured audio data, and interpreting the audio content for each captured audio data. The method further includes the steps of identifying a selected TV program by the user; determining a correlation level between the interpreted video and audio content of the selected TV program and the interpreted video and audio content of each of the other TV programs; determining TV programs other than the selected TV program; and building and presenting a mosaic display on a TV display including the selected TV program.