Method and device for video-signal processing, transmitter, and corresponding computer program product

A non-transitory computer-readable medium containing contents that cause a processing device to perform a method, the method comprising:

• identifying regions of interest in a video signal, by, identifying fast motion regions;
• identifying slow motion regions having low-level characteristics;
• comparing an area of the identified slow motion regions having low-level characteristics to a threshold area;
• when the area of the identified slow motion regions having low-level characteristics is larger than the threshold area, identifying regions of interest as a function of the fast motion regions and the slow motion regions having low-level characteristics;
• and
• when the area of identified slow motion regions having low-level characteristics is smaller than the threshold area,
• identifying face regions;
• and
• identifying regions of interest as a function of the fast motion regions and the face regions.

A device, comprising:

• a face region identifier configured to identify face regions in a video signal;
• a motion region detector configured to identify fast motion regions and slow motion regions in the video signal;
• a low-level characteristics region identifier configured to identify regions with low-level characteristics based on at least one of a contrast analysis and a skin tone analysis;
• an area calculator configured to determine an area of slow motion regions having low-level characteristics;
• a comparator configured to compare the determined area to a threshold area;
• and
• a region of interest identifier configured to:
• when the determined area is larger than the threshold area, identify a set of regions of interest as a function of identified fast motion regions and identified slow motion regions having low level characteristics;
• and
• when the determined area is smaller than the threshold area, identify the set of regions of interest as a function of identified fast motion regions and identified face regions.

A system, comprising:

• means for identifying fast motion regions;
• means for identifying slow motion regions having low-level characteristics;
• means for identifying face regions;
• and
• means for identifying regions of interest in a video signal based on an area of identified slow motion regions having low-level characteristics.