81 results about "Texture space" patented technology

A technique for determining a characteristic of a face or certain other object within a scene captured in a digital image including acquiring an image and applying a linear texture model that is constructed based on a training data set and that includes a class of objects including a first subset of model components that exhibit a dependency on directional lighting variations and a second subset of model components which are independent of directional lighting variations. A fit of the model to the face or certain other object is obtained including adjusting one or more individual values of one or more of the model components of the linear texture model. Based on the obtained fit of the model to the face or certain other object in the scene, a characteristic of the face or certain other object is determined.

A system, method, and computer program product are provided for updating texture on a graphical display object with overscan. A preprocessor stage defines an overscan region representing an extension of an object surface rasterized in texture space. A texture update stage creates a dilated texture map that includes updated mapped texels for a mapped region and updated overscan texels corresponding to the overscan region, such that texture is updated in the mapped region and the overscan region. Texel-based and polygon-based preprocessor stages and texture update stages are provided.

A technique for determining a characteristic of a face or certain other object within a scene captured in a digital image including acquiring an image and applying a linear texture model that is constructed based on a training data set and that includes a class of objects including a first subset of model components that exhibit a dependency on directional lighting variations and a second subset of model components which are independent of directional lighting variations. A fit of the model to the face or certain other object is obtained including adjusting one or more individual values of one or more of the model components of the linear texture model. Based on the obtained fit of the model to the face or certain other object in the scene, a characteristic of the face or certain other object is determined.

Min-axis based mip map determination logic receives a plurality of texture space derivatives with respect to screen space for a given pixel and texel location and selects from a plurality of mip map levels a mip map level based on a min-axis without using a max-axis value and without using an amount of anisotropy. The plurality of mip map levels corresponds to mip map levels of a mip chain. The min-axis may be identified as the squares of the texture space derivatives with respect to either the x-axis or the y-axis of screen space. Selecting the mip map level based on the min-axis ensures that each texel of the selected mip map never maps to more than one pixel during texture mapping where the main texture is of sufficient resolution. Thus, using the mip map level based on the min-axis to fetch texture data from memory and render images results in few aliasing artifacts.

The invention provides texture mapping techniques that facilitate interactive painting of a three-dimensional virtual surface by a user in object space, without requiring global parameterization. The texture mapping techniques feature rendering texture for a given virtual object using a plurality of composite textures, each formed by blending collapsible texture layers. Texture coordinates in texture space are derived using information determined at the time of surface mesh generation. The invention features dynamic texture allocation and deallocation, allowing a user to interactively modify the shape of a painted, three-dimensional model. Finally, the invention features an architecture for combined graphical rendering and haptic rendering of a virtual object, allowing a user to experience force feedback during the painting of the object in object space.

The invention puts forward a line of sight estimation method based on a generative adversarial network. The method comprises the following main contents of generating a texture, generating real data and refining eyes. The method comprises the following processes that: firstly, automatically aligning a face image with the texture space of the horizontal direction and the vertical direction of a 3Dmodel; then, mapping a synthesis image to a true domain by an unpaired pixel level domain adaptive technology; thirdly, using the annotation and synthesis data of a line of sight direction to pre-train a line of sight direction estimator; and finally, in a whole mapping process, executing a refined network to keep a line of sight direction, and using a pre-training network to serve as a conversioncirculation constraint from synthesis to truth to synthesis. By use of the method, a novel adversarial training method is used, the rendered synthesis image is mapped to a vivid domain, and accurateline of sight estimation can be obtained on a practical image without using any piece of additional flag data from a true user. For the situations of extreme head gesture, blur, long distance and thelike, the method can generate line of sight estimation with robustness.

The invention provides texture mapping techniques that facilitate interactive painting of a three-dimensional virtual surface by a user in object space, without requiring global parameterization. The texture mapping techniques feature rendering texture for a given virtual object using a plurality of composite textures, each formed by blending collapsible texture layers. Texture coordinates in texture space are derived using information determined at the time of surface mesh generation. The invention features dynamic texture allocation and deallocation, allowing a user to interactively modify the shape of a painted, three-dimensional model. Finally, the invention features an architecture for combined graphical rendering and haptic rendering of a virtual object, allowing a user to experience force feedback during the painting of the object in object space.

The invention provides a texture image compression method based on geometric information of a three-dimensional model in allusion to limitations of traditional product defect detection based on a machine vision method in applicable targets and a problem how a region-of-interest of the human eyes on a texture image is solved in the compression process. According to the invention, a region-of-interest of the human eyes is on a texture image is solved by using model grid data, three-dimensional feature points representing surface detail information of the three-dimensional model after multi-solution re-gridding and mapping points thereof on a texture space are extracted according to a visual presentation mode of the texture so as to act as texture feature points, class aggregation is carried out on the feature points in the image by using a K-means clustering algorithm according to the continuity of an image space, the region-of-interest of the texture image is acquired, and the differentiation precision of the region-of-interest and the background is improved. The method provided by the invention is verified through establishing an ROI based EZW coding and decoding experiment system, and good experiment effects are acquired.

A “mesostructure renderer” uses pre-computed multi-dimensional “generalized displacement maps” (GDM) to provide real-time rendering of general non-height-field mesostructures on both open and closed surfaces of arbitrary geometry. In general, the GDM represents the distance to solid mesostructure along any ray cast from any point within a volumetric sample. Given the pre-computed GDM, the mesostructure renderer then computes mesostructure visibility jointly in object space and texture space, thereby enabling both control of texture distortion and efficient computation of texture coordinates and shadowing. Further, in one embodiment, the mesostructure renderer uses the GDM to render mesostructures with either local or global illumination as a per-pixel process using conventional computer graphics hardware to accelerate the real-time rendering of the mesostructures. Further acceleration of mesostructure rendering is achieved in another embodiment by automatically reducing the number of triangles in the rendering pipeline according to a user-specified threshold for acceptable texture distortion.

A “mesostructure renderer” uses pre-computed multi-dimensional “generalized displacement maps” (GDM) to provide real-time rendering of general non-height-field mesostructures on both open and closed surfaces of arbitrary geometry. In general, the GDM represents the distance to solid mesostructure along any ray cast from any point within a volumetric sample. Given the pre-computed GDM, the mesostructure renderer then computes mesostructure visibility jointly in object space and texture space, thereby enabling both control of texture distortion and efficient computation of texture coordinates and shadowing. Further, in one embodiment, the mesostructure renderer uses the GDM to render mesostructures with either local or global illumination as a per-pixel process using conventional computer graphics hardware to accelerate the real-time rendering of the mesostructures. Further acceleration of mesostructure rendering is achieved in another embodiment by automatically reducing the number of triangles in the rendering pipeline according to a user-specified threshold for acceptable texture distortion.

Discontinuities along texture mapped seams of three-dimensional models may be reduced by creating and sampling texture data outside of chart boundaries. When a texel center is not within a chart boundary (a group of connected triangles in texture space) a phantom face is generated that includes the texel center. Phantom texture coordinates are created for each texel center that is covered by the phantom face. The phantom texture coordinates are used to read a texture sample from another chart in texture space that is adjacent to the chart boundary in model space, producing a smooth transition across the seam.

The invention provides a method for simplifying a vertex clustering of three-dimensional models with texture constraint. The method comprises the steps of: expanding the vertexes from a geometric space to a fifth-dimensional space including texture coordinates in consideration of algorithm efficiency in the self-adaptive division processes of the vertexes; dividing the vertexes according to the positions of the vertexes in the geometric space and the texture space in the self-adaptive division processes of the vertexes; in error measurement, respectively computing geometric errors and texture errors, if the error value is more than a given threshold, splitting the current node; and in the self-adaptive division processes of the vertexes, in consideration of saving the storage resources and improving the algorithm efficiency, adopting the relatively important vertex in an original model as a clustering representative instead of computing the geometric position of a new vertex through iteration. The method for simplifying vertex clustering of three-dimensional models with texture constraint organically integrates texture error measurement and geometric simplification, realizes rapid simplification of a mass of complex three-dimensional models and supports efficient three-dimensional visualization of city virtual scenes.

A graphics processing unit is configured to map pixels of a first frame of a video stream to texels, select a subset of the texels for shading based on previously cached texels that were shaded for a second frame, and shade the subset of the texels. The graphics processing unit is also configured to cache the shaded subset of the texels with the previously cached texels and determine values for the pixels of the first frame based on the cached texels.