How to Develop 3D Models Using 2D Images

RMN Digital | February 25, 2015 | Applications | No Comments

Neurotechnology, a provider of high-precision biometric, robotics, AI and computer vision technologies, has introduced the SentiSculpt Software Development Kit (SDK) for the creation of 3D object models.

With the broad availability of 3D printers and software for virtual reality, gaming, architecture and design applications, one of the biggest challenges has been the capture of 3D objects.

This has either required expensive, specialized photographic equipment or the drawing of complex object meshes and wireframes.

SentiSculpt SDK offers an alternative: the capture of real objects in 3D using off-the-shelf cameras or smartphones. SentiSculpt uses an advanced algorithm to turn a series of 2D photos into a 3D object model that can be used for 3D printing or in a variety of software applications.

The SDK enables developers, designers, animators and integrators to incorporate 3D technology into their designs, software, games and other solutions that require realistic, virtual environments.

“SentiSculpt brings users a higher visual experience – instead of looking through a series of static, 2D images people can navigate around objects within a 3D environment,” said Dr. Vilius Matiukas, SentiSculpt project lead for Neurotechnology.

With SentiSculpt, the user takes pictures of an object to capture the different surfaces or views (the object may be anything from a small toy to a large building, a honeybee to an aircraft carrier).

SentiSculpt begins by creating a point cloud based on the input images (2-5 megapixels is the optimal resolution for photos). SentiSculpt can then be used to generate meshes, or wireframes, from the point clouds. It can also perform texture mapping and detect the relative camera position for each photo.

SentiSculpt SDK supports C, C++ and C# development environments under the 64bit Windows platform. The SDK also includes a sample that demonstrates how plugins can be created for MeshLab, an open source system for the processing and editing of unstructured 3D triangular meshes.

SentiSculpt SDK is the latest offering in Neurotechnology’s “Senti” line of AI and robotics products, including the SentiBotics mobile robotics development kit and SentiSight SDK for development of computer and robotic vision applications.

SentiSculpt SDK and the entire line of Neurotechnology products for AI, robotics, object recognition and biometric identification are available through Neurotechnology or from distributors worldwide.

Neurotechnology is a provider of high-precision software development products for AI, robotics, object recognition and biometric fingerprint, face, iris, palmprint and voice identification.