43 results about "Photon mapping" patented technology

A circuit arrangement and method provide a hybrid rendering architecture capable of interfacing a streaming geometry frontend with a physical rendering backend using a dynamic accelerated data structure (ADS) generator. The dynamic ADS generator effectively parallelizes the generation of the ADS, such that an ADS may be built using a plurality of parallel threads of execution. By doing so, both the frontend and backend rendering processes are amendable to parallelization, and enabling if so desired real time rendering using physical rendering techniques such as ray tracing and photon mapping. Furthermore, conventional streaming geometry frontends such as OpenGL and DirectX compatible frontends can readily be adapted for use with physical rendering backends, thereby enabling developers to continue to develop with known API's, yet still obtain the benefits of physical rendering techniques.

A method for distributing photons among light sources when rendering an image of a scene using photon mapping includes the steps of computing a total energy for the scene, where the scene includes a plurality of light sources, and computing an average energy of the light sources. The method also includes the steps of comparing, for each of the light sources, the actual energy of the light source to an average energy of the light source and distributing photons to each of the light sources based on the comparison.

The invention discloses a novel photon mapping-based global illumination method, and belongs to the technical field of image processing. The method of the invention mainly comprises the following steps of: (1) randomly emitting photons to a scene by a light source; (2) judging the state of the emitted photons; (3) tracing the photons by a recursive method according to the information on the surface of an object; (4) storing the photon information; and (5) rendering the scene, namely rendering the scene by combining the Monte-Carlo backward tracing method and the primary photon mapping method. By using the technical scheme of the invention, the running speed, the rendering effect, the used storage space and the like can be improved to a certain extent.

The invention discloses a photon mapping accelerating method based on point cache. The method includes the following steps: 1 photon tracking which includes emitting a certain number of photons to a scene from a light source, then tracking the moving track of the photons, recording the information of the photons in collision with an object and storing the information into a photon picture; 2 preprocessing which includes pre-computing the irradiance, storing the irradiance into the point cache, continuing computing a tinter after the irradiance is computed and storing obtained color values in the point cache mode; 3 rendering which includes conducting rendering according to a light tracking algorithm and emitting tracking reflection, refraction and diffuse reflection light at the intersection point position of light and the surface of the object according to the object surface attribute. A final gathering algorithm emits N diffuse reflection light, the diffuse reflection light intersects with the object on the scene, the color values of N sampling points in the point cache closest to the intersection point are returned, the color values are subjected to mean value computing to serve as the indirect illumination color value, and finally the indirect illumination color value plus the direct illumination color value is the final image.

The invention discloses a parallelization type progressive photon mapping method and device based on OpenCL. The parallelization type progressive photon mapping method and device are applied to the overall illumination field in the virtual reality technology. Parallelization type progressive photon mapping is achieved through the OpenCL. The parallelization type progressive photon mapping method comprises the steps that firstly, initialization is conducted, a scene model is loaded, and the OpenCL calculating parameters are initialized; secondly, parallelization is conducted on viewpoint ray tracing, photo tracing and scene rendering based on the OpenCL, working loads are designed on corresponding processors reasonably, and after a command queue is executed, a computation result is read and transmitted to a CPU; finally, data resources stored in the CPU are released through the OpenCL standard library functions. By the adoption of the parallelization type progressive photon mapping method and device based on the OpenCL, the efficiency of the progressive photon mapping algorithm can be improved remarkably, compared with a computation method that design is conducted on the CPU, the efficiency is improved by four to nine times, the transportability is high, and the rendering effect is improved to a certain extent.

The invention discloses an overall illumination computer simulation processing method based on a photon mapping image. The method is characterized by comprising that a photon is emitted from a light source to a computer virtualized scene and is tracked, and when the photon strikes the surface of a scene content non-specular surface object, the photon is stored in the photon mapping image; incidence luminous flux of any point in the computer virtualized scene and reflected luminescence radiancy information are extracted by means of statistic technique according to the photon mapping image, calculating the pixel color value, and carrying out image drawing and rendering in the scene. The overall illumination computer simulation processing method based on the photon mapping image is based on the principle of photon mapping, effect of a processed image is life-like, calculating and processing speed is high, and meshing is not needed. The overall illumination computer simulation processing method based on the photon mapping image is especially suitable for light treatment in complex scenes.

A computer graphics system is described in which a new type of entity, referred to as a “phenomenon,” can be created, instantiated and used in rendering an image of a scene. A phenomenon is an encapsulated shader DAG comprising one or more nodes each comprising a shader, or an encapsulated set of such DAGs which are interconnected so as to cooperate, which are instantiated and attached to entities in the scene which are created during the scene definition process to define diverse types of features of a scene, including color and textural features of surfaces of objects in the scene, characteristics of volumes and geometries in the scene, features of light sources illuminating the scene, features of simulated cameras will be simulated during rendering, and numerous other features which are useful in rendering. Phenomena selected for use by an operator in connection with a scene may be predefined, or they may be constructed from base shader nodes by an operator using a phenomenon creator. The phenomenon creator ensures that phenomena are constructed so that the shaders in the DAG or cooperating DAGs can correctly cooperate during rendering of an image of the scene. Prior to being attached to a scene, a phenomenon is instantiated by providing values, or functions which are used to define the values, for each of the phenomenon's parameters, using a phenomenon editor. The phenomenon editor allows the operator to view the effects produced by various settings for the parameter values which are selected. During scene image generation, a scene image generator operates in a series of phases, including a including a preprocessing phase, a rendering phase and a post-processing phase. During a pre-processing phase, the scene image generator can perform pre-processing operations, such as shadow and photon mapping, multiple inheritance resolution, and the like. The scene image generator may perform pre-processing operations if, for example, a phenomenon attached to the scene includes a geometry shader to generate geometry defined thereby for the scene. During the rendering phase, the scene image generator renders the image. During the post-processing phase, the scene image generator may perform post-processing operations if, for example, a phenomenon attached to the scene includes a shader that defines post-processing operations.

Provided is an image processing apparatus for performing photon mapping, and the image processing apparatus may perform ray tracing for photon mapping, sample a ray space based on a result of the ray tracing, and perform pseudo photon mapping using the sampled ray space.

The invention discloses an approximate drawing method for a 3D (three-dimensional) virtual scene comprising a surface caustic effect generated by mirror reflection, belonging to the technical field of drawing of a real three-dimensional virtual scene. In the method, the 3D virtual scene comprising the surface caustic effect is drawn in three drawing steps. The approximate drawing method comprisesthe following steps of: 1, putting a virtual camera on a point light source, drawing a mirror reflection object surface in the 3D virtual scene, and creating virtual focus lamps to simulate secondaryillumination generated by mirror reflection light rays; 2, putting the virtual camera on eyes, and drawing a diffused reflection object surface of the 3D virtual scene under the irradiation of all virtual focus lamps to obtain a surface caustic drawing effect; and 3, putting the virtual camera on eyes, drawing all object surfaces of the 3D virtual scene under the irradiation of the point light source, and combining drawing results of the second step and the third step to obtain a final drawing result of the 3D virtual scene. The drawing time of the method is only equivalent to 10-15 percent of that of a photon mapping method.

The invention discloses a photon mapping parallel method for an MIC (intel many integrated core) framework coprocessor. The photon mapping parallel method comprises the following steps: a photon tracking beginning stage, namely uploading scene data to the MIC framework coprocessor end; a photon light ray intersection phase, namely carrying out intersection computation on one ling ray in each thread each time, and feeding back light ray intersection point information; a photon map generating stage, namely executing different photon impact behaviors according to the light ray intersection point information, storing impacted photons into a photon map until all photon ray beams are tracked and generating an intact photon map; a rendering beginning stage, namely uploading the photon map, organizing a plurality of groups of rendering points, and carrying out hierarchical clustering on each group of rendering points; a nearest photon researching stage, namely uploading clustered rendering points, calculating the nearest photon of each rendering point in a rendering point cluster, and determining emergent radiation intensity of each rendering point; and an image generating stage, clustered calculating the colors of the rendering points, and returning the colors of each group of rendering points to a screen space to form a final image.

The invention relates to a photon mapping rendering method and a system based on a neural network. The method comprises the following steps: 1) adopting a photon mapping method to generate k color effect images with different rendering quality; 2) inputting a plurality of training set composed of k color effect images with different rendering quality into a neural network for training to obtain aneural network model; 3) synthesizing a global lighting render image based on that neural network model. Compared with the existing progressive photon mapping method, the invention requires only the intermediate rough results produced by several iterative photon mappings, The neural network proposed by the invention can be used to infer the rendering result of synthesizing the convergence of the progressive photon mapping, thereby solving the problem that the traditional progressive photon mapping method needs a large number of iterations which are very time-consuming to generate a more idealrendering result.

The invention discloses a method for drawing surface caustic effect of a 3-dimesnioanl (3D) virtual scene generated by smooth surface refraction, and belongs to the technical field of real 3D virtual scene drawing. At present, the surface caustic effect of the 3D virtual scene is drawn generally by using a photon mapping algorithm. The photon mapping algorithm requires for tracking and computing a large amount of photons emitted by a light source so as to seriously reduce the drawing efficiency of the surface caustic effect of the 3D virtual scene. The method comprises the following steps of:creating all caustic illuminants in the 3D virtual scene generated by smooth surface refraction, and storing the caustic illuminants into corresponding data structures; and when the 3D virtual scene is drawn, judging whether the contribution of the caustic illuminants needs to be added into the illumination value of a scenic spot to be drawn by calculating the position relationship between the scenic spot to be drawn and the caustic illuminants, and finally implementing drawing of the surface caustic effect. The method can be easily integrated into a ray tracing algorithm framework, and can remarkably improve the third dimension of 3D virtual scene drawing.

The invention discloses a random asymptotic photon mapping image denoising method and system based on a neural network, and the method comprises the steps: obtaining a three-dimensional scene, and generating a to-be-denoised image based on random asymptotic photon mapping, wherein the to-be-denoised image comprises a rendering image based on global photons and a rendering image based on focal astigmatism; inputting the rendering image based on the global photons into a pre-trained first multiple residual neural network, and outputting a global photon denoising image; inputting the rendered image based on the focal astigmatism photons into a pre-trained second multiple residual neural network, and outputting a focal astigmatism photon denoising image; and synthesizing the global photon denoising image and the focal astigmatism photon denoising image to obtain a final rendering image.

The invention discloses a random asymptotic photon mapping image rendering method and system based on photon resampling, and the method comprises the steps: transmitting first light from a viewpoint,and recording the position of a collision point between the first light and a scene object; emitting second light from the light source to obtain a photon graph; dividing the photon graph into a plurality of continuous image sub-blocks, and calculating a collision point distance error value and a photon number error value in each image sub-block; in the coloring stage, traversing collision points;during first execution, carrying out photon sampling on each collision point according to a set radius; calculating a plurality of new sampling radiuses of each collision point according to the collision point distance error value and the photon number error value during non-first execution, and resampling photons in the new sampling radiuses of the collision point in the current image sub-blockby utilizing the plurality of new sampling radiuses; and when photon resampling is carried out on all the collision points, obtaining a rendered image, and returning to the coloring stage until the set number of iterations is met, so as to obtain a rendered image.

A photon map clustering method based on adaptive line-of-sight division combines a photon map ray calculation method of pre-buffering and aggregation replacement with a photon map clustering method ofadaptive line-of-sight division to achieve the purpose of optimizing a photon mapping algorithm, The specific steps are as follows: (1) in the ray tracing stage, acquiring the scene and acquiring thephoton map in the ray tracing stage; (2) Adaptive line-of-sight division stage, which divides the photons into three grades according to the adaptive line-of-sight in the photon-dense region; (3) inthe pre-buffering stage, sampling the photons after clustering to obtain the estimated radiance value and pre-buffering it into a file; (4) In the focusing rendering stage, converging the intersectionvalue, caustic photon value and direct illumination value to optimize the photon mapping and improve the image rendering efficiency and realism.

The invention relates to a deviation control method in photon mapping. The method comprises the following steps: 1, according to an estimated radius of an intersection point generated when light intersects with a scene, and collecting photons nearby the intersection point by use of a KNN method; 2, performing visibility determining on the collected photons so as to reject impure photons; 3, performing topology deviation control on the collected photons through depth continuity detection; 4, according to a normal topology structure of a surface which the collected photons are attached to, performing geometric continuity detection on the surface of the intersection point so as to control impurity deviations; 5, carrying out edge deviation control on the collected photons through convex surface examination; and 6, carrying out radiation brightness estimation by use of a radiation brightness calculation formula by taking residual photons as correct and effective photons. Then an obtained radiation brightness calculation value is converted into a color value, and thus a final highly realistic image is synthesized. The method provided by the invention can effectively eliminate the problem of deviation and noise which are generated by inappropriate collection of the photons.