The program also has difficulty picking out edges of fuzzy materials, or objects with a lot of shadows or a soft background, as is the case with the aforementioned Photoshop reference. Objects that aren’t completely visible can’t be modeled, and objects that suddenly change the type of shape involved (such as a toothpaste container) don’t build proper 3D models. Like any newly developed program, the process isn’t perfect. The computer reshapes the component to fit the image of the object in the photograph as well as to satisfy various inferred geometric constraints imposed by its global 3D structure. In our interface, three strokes are used to generate a 3D component that snaps to the shape’s outline in the photograph, where each stroke defines one dimension of the component. 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 explains how humans can help the machines:
As anyone who has ever used Photoshop could testify, computers aren’t great at grabbing the proper information from pictures without help. Dubbed 3-Sweep by the research team, the system uses both computer and user generated data to build 3D objects. The cooperative research has developed a method of pulling 3D designs from 2D images that has potential for improving both current and future 3D design.
0 kommentar(er)
