this is for holding javascript data
Paul St-Aubin edited Methodology Processing.tex
almost 10 years ago
Commit id: 16dfa07721ce75d01f3dfffce1385016ed7c44bf
deletions | additions
diff --git a/Methodology Processing.tex b/Methodology Processing.tex
index 2cc3f83..ba914e5 100644
--- a/Methodology Processing.tex
+++ b/Methodology Processing.tex
...
In the current iteration of the software, and with today's multi-core prossesors, tasks are highly parralelisable. Feature tracking and trajectory analysis can be performed on multiple video sequences at a time, typically cut up into 20 minute or 1 hour segments, in parralel on a single mid-to-high-performance machine, or on a computer cluster. With parrallel processing of video sequences on a single computer, memory becomes the main bottleneck; 32 GB or more of memory is highly recommended on a multi-core machine to take full advantage of up to 8 threads. Alternatively, the large majority of calculation tasks can be parrallelised at the observation level as they are independant events.
Feature tracking is written in C++ for performance, while
the majority of trajectory analysis is written in Python for extensibility. Where possible, expensive
trajectory analysis calculations make use of Python wrappers for pre-compiled computation libraries.