From 4ee9208dfe9685b0e1c881debde62d578451353b Mon Sep 17 00:00:00 2001 From: Martin Thoma Date: Sat, 11 Feb 2017 10:34:44 +0100 Subject: [PATCH] Fix typo (larger dimension, not smaller) See https://github.com/MartinThoma/HASY/issues/2 --- publications/hasy/content.tex | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/publications/hasy/content.tex b/publications/hasy/content.tex index 8fe7951..388e4a0 100644 --- a/publications/hasy/content.tex +++ b/publications/hasy/content.tex @@ -56,7 +56,7 @@ All elements of the data set are called \textit{recordings} in the following. \dbName{} is derived from the HWRT dataset which was first used and described in~\cite{Thoma:2014}. HWRT is an on-line recognition dataset, meaning it does not contain the handwritten symbols as images, but as point-sequences. Hence -HWRT contains strictly more information than \dbName. The smaller dimension +HWRT contains strictly more information than \dbName. The larger dimension of each recordings bounding box was scaled to be \SI{32}{\pixel}. The image was then centered within the $\SI{32}{\pixel} \times \SI{32}{\pixel}$ bounding box.