Common Vision Blox - Tools



CVB GPU Processing

Use the graphic card to increase image processing speed

With CVB GPU Processing developers can take advantage of the immense power of the GPU to increase image processing speed. The GPU processing found in Common Vision Blox is based on Microsoft's DirectX High Level Shader Language (HLSL), which means that it is fully hardware independent, allowing migration between different graphics cards and card vendors.

The CVB GPU Processing tool offers several advantages

  • Frees the programmer from complex DirectX programming
  • Takes care of the memory management on the host PC and the graphic card
  • Handles the data transfer between host and graphic card
  • Allows image access before and after GPU processing
  • Processes image data at full camera speeds
  • Supports all common acquisition technologies
  • User has full control of what algorithm runs where

CVB GPU-Processing for increased processing speed

Perfect combination of GPU and CPU

As both types of processors have their individual strengths, it is the perfect teamwork of GPU and CPU that makes the difference. In addition it has to be noted that not every algorithm is suitable for transfer to the GPU and of course the acceleration has to offset the additional image transfer between host and graphic card. The big advantage of CVB however is, that the developer has full control of what algorithm runs where and when. It supports pipeline processing of several algorithms in a sequence on the image before it is transferred back to the host memory. Additional algorithms can be added and removed from the pipeline dynamically when processing.


  • Uses the graphic card as an additional, independent CPU
  • Works asynchronously to acquisition
  • Allows pipelining of multiple processing steps for maximum performance and minimum transport time
  • Based on High Level Shader Language (part of Microsoft DirectX)
  • Includes predefined modules for standard image processing tasks
  • User algorithms can be easily developed and deployed using HLSL
  • Manufacturer independent