Methods and apparatus to monitor a split screen media presentation

A non-transitory computer readable storage medium having stored thereon instructions that, when executed by a processor of a computing system, cause performance of:

• accessing one or more video signatures generated based on at least one of multiple images corresponding to video output from a media device;
• accessing audio monitoring data extracted from audio output from the media device, wherein the audio monitoring data includes at least one of an audio signature or an audio watermark generated using the audio output;
• matching the audio monitoring data to reference audio monitoring data corresponding to first media;
• determining that the one or more video signatures do not correspond to the first media;
• in response to determining that the one or more video signatures do not correspond to the first media, assessing that the video output from the media device corresponds to a video presentation format having a plurality of candidate regions-of-interest corresponding to respective sub-regions of the video output from the media device;
• and
• identifying a first one of the candidate regions-of-interest of the video presentation format that corresponds to the first media.

An audience measurement computing system comprising:

• a processor;
• memory having stored thereon computer-readable instructions that, when executed by the processor, cause performance of operations comprising:
• accessing one or more video signatures generated based on at least one of multiple images corresponding to video output from a media device;
• accessing audio monitoring data extracted from audio output from the media device, wherein the audio monitoring data includes at least one of an audio signature or an audio watermark generated using the audio output;
• matching the audio monitoring data to reference audio monitoring data corresponding to first media;
• determining that the one or more video signatures do not correspond to the first media;
• in response to determining that the one or more video signatures do not correspond to the first media, assessing that the video output from the media device corresponds to a video presentation format having a plurality of candidate regions-of-interest corresponding to respective sub-regions of the video output from the media device;
• and
• identifying a first one of the candidate regions-of-interest of the video presentation format that corresponds to the first media.