Microsoft Technology Licensing, LLC

Inria Institut National De Recherche En Informatique Et En Automatique

Universite Dé Lorraine

Sylvain Lefebvre

Hugues Herve Hoppe

An Lu

Jeremie Dumas

John Carlton Platt

Anass Lasram

Samuel Hornus

An additive manufacturing device processes voxels representing an object to be additively manufactured by determining two graphs depending on whether the voxels indicate full or empty spaces, and redefines some of the voxels indicating an empty space as full spaces in order to ensure mechanical stability.

Device for processing data for additive manufacturing

A data processing device applies pattern data to surface data on the basis of projection data, subdividing the surface data into voxels and determining the optimum projection data for each voxel according to its surroundings.

Data processing device

An image processing device, including an analyzer calculating cutting data and difference data from image data, a selector receiving data from a working node, the data including cutting data, cost data, and position data, an assembler receiving data from working nodes and from a predecessor node and calculating updated node data as a function of the cost data, a driver configured to call the analyzer with image data of an input image, call the selector with an input node and with the cutting data calculated by the analyzer, call the assembler with the successor nodes determined by the selector as working nodes and with the input node as a predecessor node, and repeatedly call the selector and the assembler using one of the updated nodes as a working node for the selector, until a condition relating to the cutting data and the position data of an updated node is satisfied.

Image synthesis device

A method and system for synthesizing texture using upsampled pixel coordinates and a multi-resolution approach. The parallel texture synthesis technique, while based on a neighborhood matching technique having order-independent texture synthesis, extends that approach in at least two areas, including efficient parallel synthesis and intuitive user control. Pixel coordinates are upsampled instead of pixel colors, thereby reducing computational complexity and expense. These upsampled pixel coordinates then are jittered to provide texture variation. The jitter is controllable, such that a user has control over several aspects of the jitter. In addition, each neighborhood-matching pass is split into several sub-passes to improve correction. Using sub-passes improves correction speed and quality. The parallel texture synthesis system and method disclosed herein is designed for implementation on a parallel processor, such as a graphics processing unit.

Parallel texture synthesis by upsampling pixel coordinates

A method and system for synthesizing texture using a preprocessed exemplar image and a neighborhood-matching per-pixel texture synthesis correction technique. The sub-pass correction system and method alters pixel coordinates to recreate neighborhoods similar to those in the exemplar image. In the context of parallel texture synthesis, instead of synthesizing all pixels of an image simultaneously, the sub-pass correction system and method allows neighbors to be corrected in different sub-passes. Each pixel, therefore, benefits from the correction of some of its neighbors in previous sub-passes. This reduces the required number of global correction passes to obtain good synthesis results Generally, one or more correction passes are performed, with each correction pass divided into a plurality of correction sub-passes. A number of pixel coordinates are corrected in parallel during each of the correction sub-passes.

Sub-pass correction using neighborhood matching

A dimensionality-reduced appearance space system and method that transforms an exemplar image from a traditional three-dimensional space of pixel colors to a low-dimensional Euclidean space of appearance vectors. The transformation of an exemplar is a preprocessing step, and the transformed exemplar becomes the starting point for high-quality texture synthesis. The exemplar transformation begins by computing a high-dimensional appearance vector using one or a combination of several attribute channels. These attribute channels provide additional information to further distinguish exemplar pixels from each other. These attribute channels includes spatial pixel neighborhoods, feature distance, and radiance transfer information. Dimensionality reduction is applied to the resulting high-dimensional appearance vector to generate the transformed exemplar in low-dimensional Euclidean appearance space. Because much of the information contained in the high-dimensional appearance vector is redundant or coherent, dimensionality reduction can be applied to drastically reduce the dimensionality of the appearance vector with little loss of information.

Texture synthesis using dimensionality-reduced appearance space

An anisometric texture synthesis system and method for generating anisometric textures having a similar visual appearance as a given exemplar, but with varying orientation and scale. This variation is achieved by modifying the upsampling and correaction processes of the texture synthesis technique using a Jacobian field. The modified correaction process includes accessing only immediate neighbors of a pixel instead of non-local pixels. This constraint that only immediate neighbors be used also allows the generation of seamless anisometric surface textures. This is achieved by using indireaction maps containing indirection pointers that are used to jump from a set of pixels outside the boundary of a texture atlas chart to another chart. The system and method also includes an anisometric synthesis magnification technique that uses a Jacobian field to modify the magnification step of a synthesis magnification scheme and account for anisometry.

Anisometric texture synthesis

A saltating sample image enhancement system and method that provides an image processing operation in which a filter considers one or one or more exact source image pixels; one or more bilinearly interpolated source image samples, where the bilinear weights are coupled to the position of the target pixel relative to the source pixels; and (optionally) one or more linearly interpolated source image samples, where the linear weights are coupled to the position of the target pixel relative to the source pixels. The filter can construct a spatially continuous image statistic.

Image processing using saltating samples

An adaptive texture regeneration method and system for generating a sequence of images over time (an animation sequence) that gives the appearance of texture flowing over a surface. The adaptive texture regeneration method and system helps keep synthesized texture flow over a surface from becoming so distorted such that it no longer resembles the original exemplar. This is achieved in part by using pixel coordinates instead of colors. By using pixel coordinates, distortion of the texture can be measured. Based on this distortion measurement, the texture can be adaptively regenerated if necessary. The distortion measurement of the texture is measured and compared to a distortion threshold. If the measured distortion does not exceed the threshold, then the current synthesized texture is retained. On the other hand, if the measured distortion exceeds the threshold, the current synthesized texture is regenerated.

Synthesis of advecting texture using adaptive regeneration

A multidimensional hash table is created based on a data source having sparse multidimensional data. The sparse source data is mapped into the hash table using a hash function. The hash function can be defined by accessing multidimensional values in an offset table. The offset values in the offset table can be precomputed from the static source data so as to avoid hash collisions, thus creating a perfect hash function. Additionally, the perfect hash function is designed to preserve spatial coherence of accesses, so as to improve locality of memory reference.

Perfect multidimensional spatial hashing

Parallel texture synthesis having controllable jitter

A multi-level image data structure (called a multi-level image stack) containing a single image at each level and a method for generating the same. Images at each level contain the same number of pixels. The multi-level image stack defines a set of levels emanating from an original image having a number of pixels. Each successive level of the multi-level image stack contains a single image. Each single image contains the same number of pixels as the original image. Successive levels encode progressively filtered information of the original image, but without subsampling. This retains the fine-scale resolution of the original image. A variety of filter techniques can be used, including a Gaussian filter and a box filter. The multi-level image stack is particularly well-suited for use in texture synthesis applications, where its fine-scale resolution at all levels has the advantage of reducing artifacts and spatial quantization.

Multi-level image stack of filtered images

An indirection texture magnification system and method for producing high-resolution indirection texture results. The system and method uses an indirection texture, designed for use with a low-resolution texture image, and a high-resolution texture image, which is a higher-resolution version of the low-resolution texture image. The indirection texture magnification system and method re-interprets an indirection texture computed for a low-resolution image so that a higher-resolution image can be used with the same indirection texture. This generates additional samples and allows the generation of a magnified, high-resolution indirection texture result. The indirection texture magnification system and method takes three or more neighboring pixel coordinates stored in an indirection texture and offsets those pixel coordinates in order to access the higher-resolution image with an increased precision. Colors obtained from adding this offset are combined to obtain a color for a corresponding pixel at a certain location in the magnified, high-resolution indirection texture result.

Growing community of inventors

Bellevue, WA, United States of America

Sylvain Lefebvre