this is for holding javascript data
Paul St-Aubin edited Methodology Processing.tex
almost 10 years ago
Commit id: 55c0ac06a35290f3076015f7b34c01734a7aabef
deletions | additions
diff --git a/Methodology Processing.tex b/Methodology Processing.tex
index 0e3482b..07112d4 100644
--- a/Methodology Processing.tex
+++ b/Methodology Processing.tex
...
\subsection{Processing}
Real-time analysis is not an explicit goal of this technology as its intended use is primarily for research. However, performance is a serious consideration if, for no other reason than to
ensures ensure that processing remains affordable and doesn't fall behind data collection.
In any case, some tasks benefit from as much prior data as possible, particularly machine learning tasks.
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, typically cut up into 20 minute or 1 hour segments, in parralel on a single mid-to-high-performance machine. With parralel
processing, processing of video sequences, memory becomes the main
bottleneck. Most bottleneck; 32 GB or more of
the analysis memory is
conducted highly recommended on a
pair multi-core machine. Alternativly, the large majority of
dedicated consumer-grade high-performance machines. calculation tasks can be parralelised at observation level as they are independant events.