Eye-Tracked 3D Display Viewpoint Conversion for Low Crosstalk

The patent extracts and eliminates harmful crosstalk by identifying adjacent viewpoint images that cause interference and removing them from the display process. The control unit determines which adjacent viewpoint images need to be removed based on eye position and viewpoint separation, thereby eliminating the harmful factor while preserving the desired 3D image quality.

The patent applies local quality by differentiating the treatment of different viewpoint images based on their spatial relationship to the target eye position. Adjacent viewpoint images are selectively removed or adjusted only in regions where they cause crosstalk, while other viewpoint images remain unaffected. This localized approach maintains image quality where needed without unnecessarily altering the entire display content.

The patent extracts only the necessary sub-pixels that need to be activated for displaying the current viewpoint image to the target eye. By removing unnecessary sub-pixel activation based on eye position and viewpoint separation calculations, the system reduces power consumption while maintaining complete image display for the relevant viewing region.

The patent applies partial action by activating only the necessary subset of sub-pixels required for the current viewpoint display rather than all sub-pixels. This partial activation approach ensures sufficient image quality for the target eye position while avoiding the excessive energy consumption that would result from activating the entire display panel.