deletions | additions       diff --git a/untitled.tex b/untitled.tex  index a2e0039..90bf0d0 100644  --- a/untitled.tex  +++ b/untitled.tex 
...
\textit{Oh, an empty article!}   You can get started by \textbf{double clicking} this text block We partition the video volume into $C^N$ non-overlapping regions using GBH segmentation. The segmentation is based on the appearance  and begin editing. You can also click motion similarity between local regions. Each segment $c_i \in C^N$ is comprised of arbitrary shape & sized point cloud $x_i={x^0_i, x^1_i, ...., x^P_i}$ in video volume space.   The practical challenge is to represent segment $c_i$ efficiently without comprimising on  the \textbf{Insert} button below memory and accuracy. Because it is difficult  to add new block elements. Or you fit regular structure such as 3D bounding box or ellipsiod. So we came up with solution to divide the video into regular $m \times m \times m$ grids and construct the representation based on such structure. It does reduce the memory load by $m^3$ times. Also such grids  can \textbf{drag be constructed to represent arbitrary shape  and drop an image} right onto this text. Happy writing! sized 3D regions.