17 February 2015
NTT (Chiyoda-ku, Tokyo; Hiroo Unoura, CEO) has developed a new light projection technique named ‘HenGenTou’(Deformation Lamps), which can add a variety of realistic movement impressions to static objects. Based on an illusion-based animation technique, HenGenTou produces completely novel visual experiences wherein physically static objects appear to move, deform, or flutter. Given its ability to provide original dynamic impressions with traditional printed photos, painted pictures and written texts, HenGenTou has applications in a wide range of fields, including advertising, interior design, art, and entertainment.
“What is happening? Trees in a landscape oil painting are fluttering in the wind, and a man in a portrait suddenly starts speaking. This cannot be real.” What if there were some way to produce such fantastic experiences.
A promising method may be projection mapping, a cutting-edge technology that can effectively changes visual impressions of a real object, by way of projecting a well-designed pattern onto the object’s surface as a screen. Conventional projection mapping techniques, however, cannot produce movements of static objects per se, since they “paint” a new color and texture on the object’s surface without preserving the original surface’s appearance. In addition, although the goal of conventional projection mapping is to produce a physically correct image on the object’s surface, it seems theoretically impossible to make a static picture move in a physically correct way only by image projection.
NTT Communication Science Laboratories has developed a novel type of projection mapping named ‘HenGenTou’ (Deformation Lamps). HenGenTou can add a variety of dynamic impressions ranging from natural liquid flows to facial expressions to a printed image or other static materials. This capability is a result of our long-term scientific research on human information processing, in particular visual processing for natural movements.
HenGenTou makes use of visual illusions to add a movie-like dynamic appearance to a two-dimensional (2D) static object, such as fluttering flames and blowing wind. It can apparently do this even for three-dimensional (3D) objects, although the viewing condition is restricted.
Since HenGenTou can broaden the expressiveness and increase the salience of almost any static object, it will provide a new methodology for image expression in a wide range of fields.
HenGenTou can add motion impressions to static signage, such as a printed poster or a hand-written signboard, to emphasize the main message of an advertisement.
HenGenTou can induce illusory movements on the wall or floor and produce a variety of visual effects in rooms: e.g., an undulating water flow or rising heat waves.
HenGenTou can animate printed characters to make them look much more resplendent or more bizzare than the original static ones. While HenGenTou is good at adding motion impressions, conventional projection mapping is good at changing the color and texture of 3D objects. Using HenGenTou in combination with the conventional technique, artists and creators can produce novel material expressions that have never been seen before.
URL of demo movie: https://www.youtube.com/watch?v=wIHzWJm5398
HenGenTou can add realistic motion impressions to photographs, paintings, and walls and produce visual experiences that no one has ever had before.
Our technique can be applied to 3D objects. For complex 3D objects to which applying projection is difficult, superimposing motion with a transmissive LCD is a promising alternative.
By projection, HenGenTou adds only a gray-scale motion pattern to a static color picture. Since the color pattern of the picture does not move, the resulting pattern is an incorrect pattern containing many inconsistencies. However, it appears to be a correct colorful movie to human observers.
The human brain analyzes an input image separately for color, form, and motion components, and then binds them together into a coherent visual representation. When a person views the pattern produced by HenGenTou, the brain receives color and form information from the static picture, but it receives motion information from the projected pattern. Since the color and form components are stationary, they are spatially misaligned with the motion component. However, when the brain binds color, form, and motion to restore a coherent visual scene, it automatically corrects small misalignments across visual attributes. Thanks to this correction process, human observers perceive the pattern produced by HenGenTou as if the color and form components of the static picture are moving along with the projected monochrome movements. The basic idea of HenGenTou arose from the long-term investigation of human visual perception of motion and material at NTT Communication Science Laboratories.
A representative projection mapping technique, known as Shaper Lamps, projects a pattern that completely alters the original object’s appearance. On the other hand, HenGenTou only adds image movements while keeping the original appearance almost intact. While Shader Lamps is mainly used under dark illumination where the projected target is not plainly visible, HenGenTou is mainly used under bright illumination where the projected target is naturally visible. Since Shader Lamps produces a physically correct appearance on the surface of 3D objects, it requires complex CG-based computation for rendering the projected image. On the other hand, since HenGenTou produces perceptually acceptable appearances on 2D surfaces, the computation for rendering the projected image is not nearly as complex. As such, HenGenTou can produce visual experiences different from those produced by conventional projection mapping using a much simpler algorithm.
Nippon Telegraph and Telephone Corporation
Science and Core Technology Laboratory Group, Public Relations
NTT Has Instituted a Logo to Represent R&D Activities.
Information is current as of the date of issue of the individual press release.
Please be advised that information may be outdated after that point.