Vision-Space replaces picture-space: Whilst the transformation of digital media is a technical exercise, the reason why we are doing it and why it ‘works’ as a visual stimuli are complex and raise fundamental questions.

Buried very deep in most our minds is the apparent link between what we see and what the camera records. Some vision scientist are still happy to replace our real-time observation of 3D settings with photographic records of similar scenes in experimental situations. But our projection of vision does not just record ‘what’ is in the world, but also ‘how’ we see the world. We see the 3D world differently to the way we see the camera record it. Exposure to 2D media recorded by a camera will interact with us in a different way to exposure to the real 3D setting. The two are not interchangeable. Experimentation using photographs as stimuli tell us how we respond to 2D picture space media - not how we compose and use data from real 3D settings. As with some visual artists, some vision scientists are also fully sensitive to this. We are now in the position where joint investigative research could begin to explore this differential scientifically. The structure of Vision-Space draws on both the careful records of key visual artists and the insightful investigations of vision scientists.

Vision-Space raises questions about some very fundamental assumptions made about vision. If the structure of photographic representation is not seen in vision (even monocularly), then Vision-Space raises a challenge to the supremacy of binocular stereo enhancements as the being the most authentic and effective way to achieve 3D representations. 3D stereoscopic representations come from two impressions of the world recorded by the optics of the camera, not two monocular projections of the world seen in vision. Vision-Space identifies that if we want to replicate the full range of judgements we make from the real 3D world from a mechanically recovered photographic representation, this record will need to be manipulated to conform to the structure of monocular visual projection.

Recent developments in the field of sound engineering are coming up with similarities. Head Related Transfer Functions (HRTF) build into sound recordings aspects of ‘how’ we hear sound. By so doing, they give the listener additional spatial information about the source of the sound being played. We believe that there is a very strong linkage here, and that there is great potential in working further on the similarities between the spatial saliency of Vision-Space and HRTF. There are background papers that explore Vision-Space at a theoretical level available from Atelier Vision. It is likely that a MNDA will need to be in place as these papers contain commercially sensitive information.

There are background papers that explore Vision-Space at a theoretical level available from Atelier Vision. It is likely that a MNDA will need to be in place as these papers contain commercially sensitive information.

John Jupe Glasgow Talk - 2.7 MB