Volume rendering processing distribution in a graphics processing unit

• US 8,018,454 B2
• Filed: 09/20/2007
• Issued: 09/13/2011
• Est. Priority Date: 05/17/2004
• Status: Active Grant

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Brabec, Stefan, Sumanaweera, Thilaka S.
• Primary Examiner(s): NGUYEN, PHU K
• Application Number:
  US11/903,356
• Publication Number:
  US 20080018642A1
• Time in Patent Office: 1,454 Days
• Field of Search: 345/419, 345/420, 345/424, 345/426, 345/502, 600/407
• US Class Current: 345/419
• CPC Class Codes: G06T 15/04 Texture mapping
• CPC Class Codes: G06T 15/08 Volume rendering
• CPC Class Codes: G06T 19/00 Manipulating 3D models or images for computer graphics
• CPC Class Codes: G06T 2219/008 Cut plane or projection plane definition

Abstract

An intersection of a cut plane with a proxy geometry representing a scan volume is determined with a processor. The intersection is simplified, such as identifying a quadrilateral or triangle most closely enclosing the intersection. The vertex processor of a GPU deforms a reference grid and determines Cartesian coordinates and the texture coordinates for grid points of the reference grid as a function of the input intersection. The vertex processor provides coordinates for data for subsets of cut planes. The fragment processor inputs the texture coordinates and retrieves the data from the texture memory. The data is blended. The blended subsets are then blended together in the frame buffer of the GPU.

Claims

1. A method for blending during volume rendering with a graphics processing unit, the method comprising:

determining a plurality of cut planes based on a viewing direction, the cut planes being generally parallel with each other and arranged along the viewing direction;

compositing, with a fragment processor of the graphics processing unit, first and second data for first and second subsets, respectively, of the cut planes, the compositing providing composited first and second data, each of the composited first data and the composited second data being a combination or selection of data for multiple cut planes of the respective first and second subsets, the combination or selecting reducing an amount of data, the composited first data representing pixel locations of a single image and the composited second data representing the same pixel locations of the single image but at different depths relative to the viewing direction; and

compositing the composited first data with the composited second data by combination or selection to form the single image representing the pixel locations for the viewing direction.

2. The method of claim 1 wherein compositing first and second data comprises compositing first data for the first subset of three or four cut planes and compositing second data for the second subset of three or four cut planes, the cut planes of the first subset different than any cut plane of the second subset.

3. The method of claim 1 wherein compositing with the fragment processor comprises compositing prior to storage in a frame buffer and wherein compositing the composited first and second data comprises compositing from the frame buffer.

4. The method of claim 1 wherein compositing comprises blending color values and alpha values.

5. The method of claim 1 further comprising:

receiving by the fragment processor texture coordinates for each of the cut planes, the texture coordinates for the cut planes of the first subset received substantially simultaneously, the texture coordinates for the cut planes of the second subset received substantially simultaneously and the texture coordinates for the cut planes of the first subset received prior to the texture coordinates for the cut planes of the second subset; and

obtaining the first and second data as a function of texture coordinates for each of the cut planes.

6. The method of claim 1 wherein, for generating a three dimensional representation, data is read from a frame buffer half or fewer a number of times as a number of cut planes for the three dimensional representation.

7. The method of claim 1 wherein said method is used for 3-dimensional cardiology or radiology.

8. The method of claim 1 wherein compositing the first and second data comprises forming first and second data sets as a function of the data for all of the cut planes of the first and second subsets, respectively.

9. The method of claim 1 wherein compositing the first and second data comprises adding data from all the cut planes of the first and second subsets, respectively.

10. The method of claim 1 wherein compositing the first and second data comprises selecting data as a function of all the cut planes of the first and second subsets, respectively.

11. The method of claim 1 wherein the compositing the first and second data comprises compositing in a memory other than a frame buffer of a graphics processing unit.

12. The method of claim 1 further comprising processing, for each cut plane, a planar reference grid of intersecting lines in a Cartesian coordinate space with a vertex processor of the graphics processing unit, wherein compositing first and second data comprises compositing the first and second data mapped from the processed planar reference grid.

13. A method for blending during volume rendering with a graphics processing unit, the method comprising:

receiving, by a fragment processor, texture coordinates for first and second sets of planes through a volume, the planes being stacked in depth along a viewing direction, the planes of the first and second sets corresponding to first and second data at different depths along the viewing direction;

compositing the first data from the first set of planes in the volume, the composited first data representing pixel locations in an image based on the first set;

compositing the second data from the second set of planes in the volume, the composited second data representing the pixel locations in the image based the first set, the composited second data representing different depths than the composited first data; and

compositing the composited first data with the composited second data, the image having values for the pixel locations that are a function of the composited first data and the composited second data.