3D content generation for autostereoscopic displays from live-action stereoscopic 3D
Glasses-free 3D using autostereoscopic displays in digital signage is going to become a competitive alternative to conventional 2D presentations. Several companies already provide fully integrated 3D solutions for digital signage. Content creation, however, is still a major problem. While computer-generated content can easily be converted to multiview representations required for autostereoscopic displays, content creation or conversion of live-action content still remains a challenge. Fraunhofer HHI now provides an easy-to-use AFX plug-in solution for high-quality 3D content conversion from live-action stereo footage.
- Fully automatic or manually guided stereo-to-multiview conversion
- Conveniently available as Adobe AFX plug-in suite
- Use of existing stereoscopic 3D content to drive autostereoscopic multiview displays
- Support of 3D Digital Signage without glasses
- Future proved for new 3D multiview video formats
- Easy adjustment of depth parameters (IO, HIT, etc.) to achieve optimal 3D viewing comfort
To generate content for auto-stereoscopic multiview displays the AFX plug-in suite of Fraunhofer HHI’s stereo-to-multiview conversion creates additional virtual camera perspectives from available stereo input. The generation of additional views is based on disparity maps estimated from the initial stereoscopic images. Disparity maps represent pixel-by-pixel information about the depth of the scene. Sophisticated confidence measures and consistency checks coupled with detection of mismatches and tricky post-filtering of disparity maps ensure robust depth calculation. With this depth information, additional virtual camera perspectives can be generated by applying depth image-based rendering (DIBR). Thus it is possible to generate an arbitrary number of views and adapt the number and position of these virtual views to the specific properties of existing and future glasses-free 3D displays.
To produce the optimal 3D viewing experience, the depth representation can be adapted individually to each auto-stereoscopic display under consideration. This includes the level of perceived depth by adjusting the inter-axial distance (IO) as well as the position of scene objects in relation to the screen plane by adjusting horizontal image translation (HIT). To this end, by using Fraunhofer HHI’s AFX plug-in suite it can be guaranteed that the presented 3D content is always in the comfortable viewing zone of a specific auto-stereoscopic 3D display.