ContextCapture exploits the power of general purpose computation on graphics processing units (GPGPU), yielding 50-times faster processing for some operations (image interpolation, rasterization, z-buffering). It also uses multi-core computing to speed up some CPU-intensive parts of the algorithms.

ContextCapture can process 10 to 20 gigapixels per day and per ContextCapture Engine on average, depending on the hardware configuration, for the production of a textured 3D mesh with Extra precision. For input point cloud datasets, ContextCapture can process about 250 million of points per day and per ContextCapture Engine.

You can dramatically reduce processing time with grid computing, simply by running multiple ContextCapture engines on several computers, and associate them to a same job queue.

Example: for vertical + 4-oblique aerial datasets with a ground resolution of 10-15 cm and a typical overlap, we have observed an average production rate of 30-50 km² per day on a cluster of 4 ContextCapture Engines.

Regarding memory use, one ContextCapture Engine with 8 GB of RAM can handle up to 1 gigapixel of input data and 10 million output triangles in one job.