Merge branch 'main' of github.com:w-m/3dgs-compression-survey

w-m · Aug 28, 2024 · 659a4ce · 659a4ce
2 parents 2f61a9a + 5a344ac
commit 659a4ce
Show file tree

Hide file tree

Showing 3 changed files with 58 additions and 2 deletions.
diff --git a/README.md b/README.md
@@ -49,3 +49,20 @@ Baseline,xx.xx,0.xxx,0.xxx,xxxxxxxx
 Include up to two consistent submethod names for your results across all files to ensure they appear in the survey table. These names will then be concatenated with the name of your approach in the table. If you prefer only the name of your approach to be displayed in the table, you can use "Baseline" as a submethod name for those results. Results without a submethod name will only be displayed in the plots.
 
 You can view an example [here](https://github.com/fraunhoferhhi/Self-Organizing-Gaussians/tree/main/results)
+
+## LaTeX Table
+
+An up-to-date LaTeX version of the survey table can be found here [here](https://github.com/w-m/3dgs-compression-survey/blob/main/data_extraction/latex/3dgs_table.tex). Feel free to copy a line from the table into your research if you want to compare to the approach. Should you wish to copy the whole table you might need the following packages and definitions in your LaTeX preamble:
+```
+\usepackage{booktabs}
+\usepackage[table]{xcolor}
+% colors for table
+\definecolor{lightred}{HTML}{FF9999}
+\definecolor{lightyellow}{HTML}{FFFF99}
+\definecolor{lightorange}{HTML}{FFCC99}
+\usepackage{makecell}
+\usepackage{adjustbox}
+% make text the same size even when its bold in a table
+\newsavebox\CBox
+\def\textBF#1{\sbox\CBox{#1}\resizebox{\wd\CBox}{\ht\CBox}{\textbf{#1}}}
+```
diff --git a/data_extraction/latex/3dgs_table.tex b/data_extraction/latex/3dgs_table.tex
@@ -0,0 +1,38 @@
+\section*{Survey Table}
+% Have a look into ../tex_templates/preamble.tex for packages and color definitions 
+
+\rowcolors{2}{gray!25}{white}
+{
+\footnotesize
+\setlength{\tabcolsep}{3pt}
+
+\begin{adjustbox}{center}
+\begin{tabular}{ll|llll|llll|llll|llll}
+\toprule
+Method & Rank & \multicolumn{4}{c|}{TanksAndTemples} & \multicolumn{4}{c|}{MipNeRF360} & \multicolumn{4}{c|}{DeepBlending} & \multicolumn{4}{c}{SyntheticNeRF} \\
+ & \tiny  & \tiny PSNR$\uparrow$ & \tiny SSIM$\uparrow$ & \tiny LPIPS$\downarrow$ & \tiny \makecell{Size \\ MB$\downarrow$} & \tiny PSNR$\uparrow$ & \tiny SSIM$\uparrow$ & \tiny LPIPS$\downarrow$ & \tiny \makecell{Size \\ MB$\downarrow$} & \tiny PSNR$\uparrow$ & \tiny SSIM$\uparrow$ & \tiny LPIPS$\downarrow$ & \tiny \makecell{Size \\ MB$\downarrow$} & \tiny PSNR$\uparrow$ & \tiny SSIM$\uparrow$ & \tiny LPIPS$\downarrow$ & \tiny \makecell{Size \\ MB$\downarrow$} \\
+\midrule
+HAC-lowrate & \cellcolor{lightred}4.6 & 24.04 & 0.846 & \cellcolor{lightyellow}0.187 & \cellcolor{lightred}8.1 & 27.53 & 0.807 & 0.238 & \cellcolor{lightorange}15.3 & 29.98 & 0.902 & 0.269 & \cellcolor{lightred}4.3 & 33.24 & 0.967 & 0.037 & \cellcolor{lightred}1.2 \\
+HAC-highrate & \cellcolor{lightred}4.6 & \cellcolor{lightyellow}24.40 & \cellcolor{lightyellow}0.853 & \cellcolor{lightorange}0.177 & 11.2 & \cellcolor{lightorange}27.77 & 0.811 & 0.230 & 21.9 & 30.34 & 0.906 & 0.258 & \cellcolor{lightyellow}6.3 & \cellcolor{lightorange}33.71 & 0.968 & 0.034 & \cellcolor{lightyellow}1.9 \\
+gsplat-1.00M & \cellcolor{lightorange}5.2 & 24.03 & \cellcolor{lightorange}0.857 & \cellcolor{lightred}0.163 & 15.4 & 27.29 & 0.811 & \cellcolor{lightyellow}0.229 & \cellcolor{lightorange}15.3 &  &  &  &  &  &  &  &  \\
+IGS-Low & \cellcolor{lightyellow}5.7 & 23.70 & 0.836 & 0.227 & \cellcolor{lightyellow}8.4 & 27.33 & 0.809 & 0.257 & \cellcolor{lightred}12.5 & \cellcolor{lightorange}30.63 & 0.904 & 0.293 & \cellcolor{lightyellow}6.3 & 33.36 & \cellcolor{lightorange}0.971 & 0.036 & \cellcolor{lightorange}1.8 \\
+IGS-High & 5.9 & 24.05 & 0.849 & 0.210 & 12.5 & 27.62 & \cellcolor{lightorange}0.819 & 0.247 & 25.4 & \cellcolor{lightred}32.33 & \cellcolor{lightred}0.924 & 0.253 & 7.7 & \cellcolor{lightred}34.18 & \cellcolor{lightred}0.975 & \cellcolor{lightorange}0.032 & 2.7 \\
+Morgenstern et al. w/o SH & 6.0 & \cellcolor{lightorange}25.27 & \cellcolor{lightorange}0.857 & 0.217 & \cellcolor{lightorange}8.2 & 27.02 & 0.803 & 0.232 & \cellcolor{lightyellow}16.7 & \cellcolor{lightyellow}30.50 & 0.908 & 0.261 & \cellcolor{lightorange}5.5 & 31.75 & 0.961 & 0.040 & 2.0 \\
+Morgenstern et al. & 7.5 & \cellcolor{lightred}25.63 & \cellcolor{lightred}0.864 & 0.208 & 21.4 & \cellcolor{lightyellow}27.64 & \cellcolor{lightyellow}0.814 & \cellcolor{lightred}0.220 & 40.3 & 30.35 & \cellcolor{lightyellow}0.909 & 0.258 & 16.8 & \cellcolor{lightyellow}33.70 & \cellcolor{lightyellow}0.969 & \cellcolor{lightred}0.031 & 4.1 \\
+Navaneet et al. 32K & 7.8 & 23.44 & 0.838 & 0.198 & 13.0 & 27.12 & 0.806 & 0.240 & 19.0 & 29.90 & 0.907 & \cellcolor{lightyellow}0.251 & 13.0 &  &  &  &  \\
+Navaneet et al. 16K & 8.0 & 23.39 & 0.836 & 0.200 & 12.0 & 27.03 & 0.804 & 0.243 & 18.0 & 29.90 & 0.906 & 0.252 & 12.0 &  &  &  &  \\
+RDO-Gaussian & 8.4 & 23.34 & 0.835 & 0.195 & 11.5 & 27.05 & 0.802 & 0.239 & 22.4 & 29.63 & 0.902 & 0.252 & 17.2 & 33.12 & 0.967 & 0.034 & 2.2 \\
+Reduced3DGS & 8.9 & 23.57 & 0.84 & 0.188 & 14.0 & 27.10 & 0.809 & \cellcolor{lightorange}0.226 & 29.0 & 29.63 & 0.902 & \cellcolor{lightred}0.249 & 18.0 &  &  &  &  \\
+Compressed3D & 10.7 & 23.32 & 0.832 & 0.194 & 17.3 & 26.98 & 0.801 & 0.238 & 28.8 & 29.38 & 0.898 & 0.253 & 25.3 & 32.94 & 0.967 & \cellcolor{lightyellow}0.033 & 3.7 \\
+Scaffold-GS & 10.9 & 23.96 & \cellcolor{lightyellow}0.853 & \cellcolor{lightorange}0.177 & 87.0 & \cellcolor{lightred}28.84 & \cellcolor{lightred}0.848 & \cellcolor{lightred}0.220 & 156.0 & 30.21 & 0.906 & 0.254 & 66.0 &  &  &  &  \\
+Compact3DGS+PP & 11.5 & 23.32 & 0.831 & 0.202 & 20.9 & 27.03 & 0.797 & 0.247 & 29.1 & 29.73 & 0.900 & 0.258 & 23.8 & 32.88 & 0.968 & 0.034 & 2.7 \\
+EAGLES & 11.9 & 23.37 & 0.84 & 0.20 & 29.0 & 27.23 & 0.81 & 0.24 & 54.0 & 29.86 & \cellcolor{lightorange}0.91 & \cellcolor{lightorange}0.25 & 52.0 &  &  &  &  \\
+Compact3DGS & 12.9 & 23.32 & 0.831 & 0.201 & 39.4 & 27.08 & 0.798 & 0.247 & 48.8 & 29.79 & 0.901 & 0.258 & 43.2 & 33.33 & 0.968 & 0.034 & 5.5 \\
+LightGaussian & 13.0 & 23.11 & 0.817 & 0.231 & 22.0 & 27.28 & 0.805 & 0.243 & 42.0 &  &  &  &  & 32.72 & 0.965 & 0.037 & 7.8 \\
+EAGLES-Small & 13.7 & 23.10 & 0.82 & 0.22 & 19.0 & 26.94 & 0.80 & 0.25 & 47.0 & 29.92 & 0.90 & \cellcolor{lightorange}0.25 & 33.0 &  &  &  &  \\
+\bottomrule
+\end{tabular}
+\end{adjustbox}
+\newline\newline
+\noindent Note: The best methods in each category are highlighted (\colorbox{lightred}{fist}, \colorbox{lightorange}{second}, \colorbox{lightyellow}{third}). The ranks represent the average rankings of the methods across all available datasets. The quality metrics PSNR, SSIM, and LPIPS are equally weighted with the model size, meaning they each contribute one-sixth to the ranks, while the size contributes half.
+}
diff --git a/data_extraction/latex/tex_templates/table.tex b/data_extraction/latex/tex_templates/table.tex
@@ -1,5 +1,6 @@
 \section*{Survey Table}
-% Your introduction content
+% Have a look into ../tex_templates/preamble.tex for packages and color definitions 
+
 \rowcolors{2}{gray!25}{white}
 {
 \footnotesize
@@ -9,5 +10,5 @@ \section*{Survey Table}
 <table>
 \end{adjustbox}
 \newline\newline
-\noindent Note: The best methods in each category are highlighted (\colorbox{lightred}{fist}, \colorbox{lightor}{second}, \colorbox{lightyellow}{third}). The ranks represent the average rankings of the methods across all available datasets. The quality metrics PSNR, SSIM, and LPIPS are equally weighted with the model size, meaning they each contribute one-sixth to the ranks, while the size contributes half.
+\noindent Note: The best methods in each category are highlighted (\colorbox{lightred}{fist}, \colorbox{lightorange}{second}, \colorbox{lightyellow}{third}). The ranks represent the average rankings of the methods across all available datasets. The quality metrics PSNR, SSIM, and LPIPS are equally weighted with the model size, meaning they each contribute one-sixth to the ranks, while the size contributes half.
 }