The Fuel-3D (Oxfordshire, UK) 3D scanning system is based on technology developed by Professor Ron Daniel at Oxford University. The company raised over $300,000 in 2013 with a Kickstarter campaign. The scanner was originally developed for medical imaging applications. Since then, Fuel-3D has raised millions in private investment while the company mission has evolved to that of adapting the technology for the broader 3D market.
A video illustrating the Scanify in operation can be found at the end of this article.
When taking an image, the scanner’s three Xenon flash bulbs fire almost simultaneously. By this means, the two cameras in the scanner acquire a series of stereoscopic 2D photographs of an object illuminated from three slightly different directions. From these photographs, image processing is used to analyze the way that the observed illumination level varies across the object with the change in lighting direction. With the addition of a corresponding map of reflectivity, this allows calculation of the direction of the normal to the surface of the object for each pixel in the image. The resulting so-called “normal map” is then integrated to provide a highly detailed 3D “range map” of the object.
The series of 2D photographs are captured in under 0.1 second. Despite this high speed, the Scanify is a handheld device and it is quite possible for the object to move during the data acquisition time period. This issue is addressed by placing a simple target with specific optical characteristics near or on the object. The system is programmed to know the size and layout of the target and, by looking for the target in the image, the scanner can accurately estimate the relative position and orientation of the scanner with respect to the target – even if the object moves.
The Scanify system tracks surface texture using a combination of stereoscopy and photometry. Stereoscopic imaging is used to acquire accurate underlying 3D shape information from the object. Photometric-stereo is used to acquire color and high-resolution 3D detail from the object.
The resulting 3D image produced by the software consists of a large number of samples, each having XYZ coordinates and material properties in 8 bit RGB.
The scanner’s operating range is between about 350 mm and 450 mm. Acquiring a larger 3D model requires that multiple scans be stitched together.
The final output resolution from Scanify varies with the distance of the scanner from the object. The scanner is capable of a resolution of approximately 350 microns.
It can be noted that, for the scanner to work well, the object must have a degree of random surface texture, either from variation in color, or from having a rough or wrinkled surface. Objects that are less suitable for scanning includes those that are monochromatic, reflective or transparent. In addition, Scanify cannot scan behind objects or around bulky protrusions.
Formats that can be output by Scanify include STL, used by many 3D printers, and OBJ and PLY for design or animation software environments.
The Scanify is currently available and sells at a price of $1,490 excluding tax. –Arthur Berman
