Microsoft Technology Licensing, LLC

Tibor Szaszi

Kristofer Noel Iverson

Emmett Walter Lalish

Vladimir Sisolak

Jan Jakubovic

Milos Bazelides

Anton Mateasik

Seth William Orr

Gheorghe Marius Gheorghescu

Martin Kusnier

Techniques are described for transforming image data, such as two dimensional (2D) or partial 3D image data (image data), into a 3D model. Upon receiving image data including color information, the image data may be segmented into a plurality of segments using the color information. At least one height value may be assigned to each of the plurality of segments based on the color information, to define 3D image data. From the 3D image data, a 3D model may be generated, for example, for visualization, modification, and/or 3D printing. In some aspects, segmenting the image data may include comparing intensity values associated with pixels and forming edges in the image data if the intensity values differ by a threshold amount. Multiple edges may be determined and connected to form one or more contour loops, whereby the contour loop(s) may be extruded to produce complete 3D image data.

2D image processing for extrusion into 3D objects

Techniques are described for generating a three dimensional (3D) object from complete or partial 3D data. Image data defining or partially defining a 3D object may be obtained. Using that data, a common plane facing surface of the 3D object may be defined that is substantially parallel to a common plane (e.g., ground plane). One or more edges of the common plane facing surface may be determined, and extended to the common plane. A bottom surface, which is bound by the one or more extended edges and is parallel with the common plane, may be generated based on the common-plane facing surface. In some aspects, defining the common plane facing surface may include segmenting the image data into a plurality of polygons, orienting at least one of the polygons to face the common plane, and discarding occluding polygons.

Technique for extruding a 3D object into a plane

Techniques are described for transforming image data, such as two dimensional (2D) or partial three dimensional (3D) image data, into a 3D model. Upon receiving image data including color information, the image data may be converted into a height map based on the color information. The height map may be used to construct an image data mesh, which forms a 3D model. In some aspects, constructing the image data mesh may include associating vertices with pixels of the image data, connecting neighboring vertices to define at least one surface, applying texture to at least one of the surfaces, generating bottom and side surfaces, and connecting the bottom and side surface(s) to the textured surface to enclose a volume within the 3D model. In some aspects, the height map may include an edge based height map, such that color distances between pixels may be used form edges from the image data.

Growing community of inventors

Bratislava, Slovakia

Tibor Szaszi