Back to the top

Main Home

Converting 2D into 3D
‘Unparalleled visual immersion, as if being there’

We are proud to present the world’s first method to convert ordinary 2D video images to true-to-nature 3D images. Using our patented TU Delft developed algorithm, the existing information in the video allows us to obtain and create the distance of objects within the video images like a human eye would.

This is the next step in video conversion – where after black-and-white to colour – now also 3D can be added to 2D images, making expensive 3D camera’s and the associated post-processing obsolete.

Already the ancient Greeks found out that in painting depth one must use the fact that everything further away is darker, and more blurred. Medieval painters forgot this, hence those paintings appear flat, but in the Renaissance the school of Caravaggio rediscovered this “claroscuro” or “clair-obscur”. The famous Flemish and Dutch painters of the 17th century studied the Italian masters and often traveled to Italy. Rembrandt was called the master of light and in his masterpiece the Nachtwacht (Nighwatch) he used all the tricks of the books known at that time to show his mastership.

Our eyes evolutionary developed with only one light source, the sun. In that situation the fact that everything further away is darker holds, only shading and shadows ruin this law. How does the brain do it and how can we get rid of them? We developed a method to filter out lighting artefacts in an image, resulting in an image that looks like London in the fog; grey, no shiny surfaces and the light comes from nowhere and everywhere. In such situations Depth from Luminance (DFL), as we call the clair-obscur method in image processing terms, holds. After this filter, we use DFL to generate a depth map, and from there we create SBS and ANA versions. Rembrandt deliberately painted two persons in front of the scene in totally white and totally black robes. He also put multiple light sources in the painting, notably on a girl in a white dress halfway in the painting. Probably he wanted to show that everything darker is further away is not a pure cue for depth. Our brains have learned to cope with these exceptions; we use contextual information too. For instance that we see or inferred where the people stand on the floor in the scene.

Use our method on an image of the Nachtwacht and you will be amazed; how good our method is, and how good Rembrandt could paint depth.

QdepQ is a startup company of the Delft University of Technology in the Netherlands, aiming at bringing 3D viewing technology into everyday life. It is our belief that, just as color TV displaced B/W TV in the past, 3D Cinema, 3D TV and auto stereoscopic (without glasses) 3D screens will eventually enter our daily life. With our patented system, generating 3D content is easy for still pictures, video, and cinema. Even transforming old analog images and films might give surprising results, as they are not compressed. Our expertise is in optics, image processing, software, and digital hardware design.

Markets & Applications

For the last 100+ years we have learned about the stories of the world through 2D pictures and videos. Finally we are able to breach through to the third dimension for all photo-real material. Thanks to our QdepQ technology, archives can be rerendered into 3D, films can become more immersive for a fraction of the normally needed 3d camera cost and viewers will have an opportunity experience reality as never before.

We extent the physical reality that 360 video can give you by adding depth to the environment shown. This allows greatness, proportion and dimensionality to take centre stage in the viewing experience.

Using our technology, we extract depth information out of photography. This data can be used for 3D-printing accurate and detailed pictures. Markets include: Education, Art, Security (badges with 3D-printed pictures), Entertainment, and Decoration.

How it works

Team

The team working daily with great passion to enable the world’s first true 3D conversion algorithm

vikko

Vikko Smit

Co-Founder

MSc Computer Science

pieter

Pieter Jonker

Co-Founder

Emeritus Professor Cognitive Robotics

Contact

Find us:
Juliana van Stolberglaan 10
2595CL The Hague
The Netherlands

info@qdepq.com