Abstract:
Stereoscopic displays like head-mounted displays (HMDs) can lead to vergence-accommodation conflict (VAC). Studies have identified zone of comfort, specifying acceptable ...Show MoreMetadata
Abstract:
Stereoscopic displays like head-mounted displays (HMDs) can lead to vergence-accommodation conflict (VAC). Studies have identified zone of comfort, specifying acceptable range of vergence and accommodation distances on given viewing conditions. Contents rendered for different viewing conditions may fall outside this zone and require adjustments for viewing comfort. However, managing this on the server side presents challenges, particularly when device-side decisions, like interpupillary distance (IPD) adjustments on HMDs, can alter viewing conditions. Thus, in this paper, we propose an efficient device-side framework for optimizing stereo contents. The method receives a bitstream with stereo video and disparity metadata. It jointly analyzes the viewing condition parameters and metadata to derive comfort zone disparity ranges and subsequently employs scale and shift operations to adjust disparities, optimizing them within the comfort zone while minimizing content disturbance. We demonstrate our method's effectiveness through human study results, highlighting its ability to enhance viewing comfort with low computational demands.
Published in: 2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR)
Date of Conference: 07-09 August 2024
Date Added to IEEE Xplore: 15 October 2024
ISBN Information: