Head motion dependent viewport region modification for omnidirectional conversational vdd

Abstract

A method for packing a frame of immersive or omnidirectional content may include: determining one or more head motion statistics for rendering and packing immersive or omnidirectional content; determining a classification of a head motion based, at least partially, on the one or more determined head motion statistics; determining resolution(s) and/or a packing scheme based on determined viewport change rate, extent, change prediction/frequency, etc., as summarized by FIG. If there is little or no head motion, the majority of a frame may be packed with high quality and resolution content of the viewport. If there is a greater amount of head motion, a smaller portion of the frame may be packed with high quality and resolution content of the viewport, and medium/low resolution content may be included with the frame that is associated with non-viewport content.