Intel Corporation

Stewart N Taylor

Scott A Krig

Michael F Fallon

Gururaj Nagendra

Ryan Metcalfe

Dmitry Krasnov

Matthew James Adiletta

Myles Wilde

Yuenian Yang

Yuh-Lin E Chang

A method and system for managing consistent data objects are included herein. The method includes detecting an operation to store a consistent data object. Additionally, the method includes detecting an attribute for the consistent data object. Furthermore, the method includes storing the consistent data object based on the attribute. In addition, the method includes determining an additional format of the consistent data object is to be stored. The method also includes generating a second consistent data object based on the additional format and storing the second consistent data object.

Managing consistent data objects

Techniques related to accelerating color conversion are discussed. Such techniques may include generating a converted color value based on an array of ordered coefficients associated with a subsection of a section of a color conversion space and input color channel value offsets within the section of the color conversion space.

Acceleration of color conversion

A flexible representation of fine grain image buffer validity is included in an image graph implementation API to provide a mechanism for a graph node developer to communicate limits of scheduling constraints imposed by a graph's source and destination images. An image graph executor may employ a producer and/or consumer node scoreboard object or function defined through the image graph implementation API to schedule work tasks to hardware resources in a more efficient manner. In some embodiments, an image processing engine is configured to begin processing completed source data blocks (e.g., image tiles) through an image graph before all upstream tasks associated with the source image are complete. In further embodiments, a destination image is to be indicated as at least partially complete so that work dependent on one more completed data blocks may begin before the destination image is fully buffered into memory.

Graph-based application programming interface architectures with producer/consumer nodes for enhanced image processing parallelism

A flexible representation of library function equivalency is included in an image graph implementation API to provide a mechanism for a graph node developer to identify functions that may be executed on heterogeneous devices to arrive at sufficiently equivalent results. An image graph executor may employ an equivalency class object or function defined through the image graph implementation API to schedule work subtasks to heterogeneous hardware resources in parallel for greater implementation efficiency.

Graph-based application programming interface architectures with equivalency classes for enhanced image processing parallelism

A flexible representation of destination-source data block dependency is included in an image graph implementation optimization API to provide a mechanism for a graph node developer to communicate limits of scheduling constraints imposed by a graph's node connections. An image graph compiler and/or image graph executor may employ the destination-source data block dependency defined through the image graph implementation API to schedule work tasks to hardware resources in a more efficient manner.

Graph-based application programming interface architectures with node-based destination-source mapping for enhanced image processing parallelism

Techniques related to rendering scanned images are discussed. Such techniques may include selectively processing segments of a scanned input image based on a neutral edge tag signal such that segments are color suppressed and edge enhanced when the neutral edge tag signal is asserted and error diffusion processed when the neutral edge tag signal is not asserted.

Rendering high quality images via micro-segmentation, error diffusion, and edge enhancement

According to some embodiments, an image pipeline controller may determine an image stream having a plurality of image primitives to be executed. Each image primitive may be, for example, associated with an image algorithm and a set of primitive attributes. The image pipeline controller may then automatically deploy the set of image primitives to an image computation fabric based at least in part on primitive attributes.

Image stream pipeline controller for deploying image primitives to a computation fabric

Systems and methods of operating a computing system may involve utilizing an graphics engine to complete a print request. Upon receiving a request to print a scheduling management application may direct an application programming interface call to the graphics engine, and the graphics engine may process a print imaging instruction.

Using a graphics engine to process print requests

Methods, systems and computer program products to improve the efficiency and computational speed of an image enhancement process. In an embodiment, information that is generated as interim results during feature extraction may be used in a segmentation and classification process and in a content adaptive enhancement process. In particular, a cleaner image that is generated during a noise removal phase of feature extraction may be used in a content adaptive enhancement process. This saves the content adaptive enhancement process from having to generate a cleaner image on its own. In addition, low-level segmentation information that is generated during a neighborhood analysis and cleanup phase of feature extraction may be used in a segmentation and classification process. This saves the segmentation and classification process from having to generate low-level segmentation information on its own.

System, method and computer program product for document image analysis using feature extraction functions

A method to automatically replace computationally intensive functions with optimized functions in managed code is disclosed. If the underlying processor has associated optimized functions, managed application code is disassembled and parsed to find computationally intensive functions. The computationally intensive functions are then replaced with optimized functions, and the application code is re-compiled if necessary.

Using optimized libraries to improve performance of deployed application code at runtime

Methods and apparatus to optimized managed application program interfaces are described herein. In an example method, a processor instruction proxy stub (PIPS) associated with one or more processor instructions is generated. Based on the PIPS, an optimized managed application program interface is generated to execute the one or more processor instructions.

Growing community of inventors

Los Altos, CA, United States of America

Stewart N Taylor