Watch this demo of a tool that can create 3D objects from a regular 2D image.
Modelling products in three dimensions can be a time-consuming process. It is easy enough for humans to identify the presence of 3D objects in 2D scenes, but extracting these with algorithms is tricky. A new system developed by a team of researchers from the Interdisciplinary Centre in Israel and Tel Aviv University aims to simplify the process by using a technique to pull out 3D shapes and objects from regular photos.
The system is simple enough to be used and understood by almost anyone. To start with, the user traces around the object, defining its three dimensions. Then the outline will automatically snap to the object in the photo.
But 3-Sweep is not just for tracing out 3D objects. The resulting model can then be edited or manipulated, and placed back into photos or 3D scenes. This might not sound like much on paper until you see the results in the video below.
While the method won't work on every photo, the initial product is designed for ease of use and making things as easy as possible for beginners. "Our biggest goal is still to help novice users to do this," said Tao Chen to Wired. "It took us some time to figure out how to make a very convenient user interface to generate this stuff."