Visual discomfort prevention: A new motion-based comfort zone

• Werner Zellinger

Depth mapping algorithms enable the adaption of the 3D scene’s depth budget to generate new virtual views of the stereoscopic video scenes. These depth budget adaptions are limited by so-called comfort zone limits. In contrast to standard approaches for comfort zones, this paper proposes a video characteristic dependent comfort zone. It is derived by motion feature extraction and data analysis of expert knowledge from film industry. In addition, a best-practice design pattern of stereographers’ is analysed, that is, aligning the most attracting objects onto the screen plane. In subjective assessments, the depth mappings of various self-captured and publicly available stereoscopic videos are compared. Results show, that the proposed approach can be used to reduce visual discomfort associated with stereoscopic videos, where the depth quantity and the subjective image quality are not deteriorated. All stereoscopic videos and subjective assessment results proposed in this paper are put online for public reference.