Disney Research, Zürich has developed a new program that can build 3D models from 2D photographs.
Using hundreds of photographic images and a specially-designed algorithm, the program can create 3D models of complex, real-life scenes for movie, TV and games, and be used to make or print high-resolution models.
Building 3D models from multiple 2D images captured from a variety of viewing positions is nothing new, but doing so for highly detailed or cluttered environments at high resolution has proved difficult because of the large amounts of data involved.
The Disney Research, Zürich team, however, developed an algorithm that can effectively leverage these amounts of data, and process them efficiently without the need to keep all of the input data in memory at one time.
The research is published in a paper called Scene Reconstruction from High Spatio-Angular Resolution Light Fields.
Many 3D models now are obtained using laser scanning. In complex, cluttered environments, however, a single laser scan misses a lot of detail because objects in the foreground can block the laser's view. Photography makes it easier to capture the scene from multiple viewpoints, but combining photographs to build a 3D model is burdensome at high resolutions.
Disney Research's algorithm first computes reliable depth estimates specifically around object boundaries instead of interior regions, by operating on individual light rays instead of image patches. More homogeneous interior regions are then processed in a fine-to-coarse procedure rather than the standard coarse-to-fine approaches. "This allows our algorithm to retain precise object contours while still ensuring smooth reconstructions in less detailed areas." writes the paper.
According to Disney Research, this new method could be used to build and create accurate reconstructions of scenes for films and video games, or the models could be printed with all the advancements in 3D printing.
The researchers demonstrated their method by photographing a number of complex outdoor and indoor scenes with a standard DSLR camera, using 100 21-megapixel-resolution images to create each 3D reconstruction. Most existing stereo reconstruction techniques have been tailored for resolutions of just 1 or 2 megapixels.
The photos were captured along a linear path; this geometry provided structure that the researchers could leverage to make processing the data more efficient. However, the researchers also generalized their approach so that it can be applied even to a set of images taken with a hand-held camera.
Posted in 3D Software
Maybe you also like:
- Make It Stand: balancing shapes for 3D printing
- MIT Develops New Ways to Streamline, Simplify 3D Printing
- Just-released Fedora 19 adds native support for 3D printers
- 3D sculpting tool Leopoly turns everyone into 3D modellers
- Microsoft adds 3D printing support to Windows 8.1, 3D printers soon in MS stores
- An Interview with Cura developer David Braam
- Create easy 3D models with Pirate 3D's Smart Objects
- 3D imaging market growing at 26.7% CAGR & to reach $9.82 billion by 2018
- Siemens donates $440 million to prepare students at YSU for advanced manufacturing
- Free Cubify Draw app turns your fingers into an instant 3D drawing tool
- Tinkercad reopens, Autodesk agrees to acquire Tinkercad
- Shape-It-Up lets designers manipulate 3D shapes using hand gestures
- 3D Systems can turn you into a Star Trek figure for $70
braunbaer wrote at 7/30/2013 2:47:26 PM:
could this be used to create superb 3d models from old disney movies like micky mouse? Drawn frames are nothin else then images are they?
lol wrote at 7/30/2013 11:49:54 AM:
Look like: ://www.youtube.com/watch?v=OnqdXbop8F0
Mouldmesh wrote at 7/29/2013 6:18:40 PM:
Looks very good, but this technique of scanning 3D objects cannot compete yet with laser or structured lighting scanning. It gives the right 3D environment, but try to print that and it will give come out with very little details (definitely not even 1cm details) Better than 123D-Catch anyway SL scanning can use a DSLR camera as well, with as many megapixels as you wish... not practicle to work on the 3D object though... unless you have a supercomputer.