As we all know, getting from 2D to 3D in various CAD packages has typically involved importing multiple sketches, centering at the origin of the XYZ, and building out from there. At the recent SIGGRAPH Asia however, a new software by the name of 3-Sweep was previewed that allows a user to extract 3D data from a single 2D photograph. Sound too good to be true? Let’s take a look.
“Our technique provides the user the means to quickly create editable 3D parts human assistance implicitly segments a complex object into its components, and positions them in space.”
-Ariel Shamir, 3-Sweep co-creator
While the software looks promising, it looks like it works best with high-contrast photographs with prominent depth-of-field. How often are you designing a product from a high-contrast photograph? Sure it’s great for manipulating a pre-existing real world object, but how will it fit into a design workflow (assuming you’re designing a ‘new’ object)? If the example of the wine glass sketch in the video is any indication of what we might see next from the team, than perhaps this just might be a true game-changer for product designers and makers. At the current moment though, it’ll be interesting to see how the software plays out in the CG and game design field where creating models of pre-existing real-world objects can be a time-consuming and repetitive task:
Regardless, it’s stuff like this that gets our hearts pumping for what the future holds for CAD. To read more on the software, be sure to head over to project co-creator Ariel Shamir’s site.
(images via 3-Sweep/Ariel Shamir)