CUDA VisionWorkbench v 126.96.36.199
CUDA VisionWorkbench (CVWB) is an application for 32 or 64 bit versions of Windows XP or Vista, used primarily for
demonstration, benchmarking and development of vision primitives implemented in CUDA. CVWB presents an
interactive UI for calling various image processing operations on 8 bit or 16 bit grayscale images of a variety of sizes.
The UI gives the user options to load images from files or generate test patterns, select enumerated GPU’s or the CPU for
processing, select # of threads for host processing, select processing bit depth, and load, re-configure and save structuring
element matrices for convolution and correlation operators. Processing operations can be triggered individually or in a
repeating loop, and a representative multi-operator sequence can also be initiated. Results are presented visually for
interactive comparison, and key metrics are automatically recorded to pre-formatted files for analysis.
Uses for the CVWB Application and DLL
1) Benchmarking and testing vision primitive functions implemented in CUDA for a GPU and C++ for a host CPU.
This may be useful, for example, for comparison of different GPU and host CPU hardware configurations, iterative
development and evaluation of functions and informal regression testing.
2) Demonstrate optimized but reasonably transparent CUDA code in the context of a whole application.
3) Provide a basic set of CUDA and C++ vision processing primitives (with plans to continue adding more).
4) Demonstrate a way to obtain a large “pipeline benefit” for multiple processing operations, with highly modular
CUDA code. Important and effective for masking host-CPU transfer time for processing pipelines.
5) Show a way to use managed, safe .Net code for application programming, while still obtaining large acceleration from
the GPU via CUDA code contained in an unmanaged DLL.
1) CVWB Operating Prerequisites
a) x86 / AMD64 CPU based system running