Restructering LaTex

data_extraction/latex/build_latex.py

@@ -21,12 +21,6 @@
 combined_markdown = ""
 section_count = 1
 
-with open("tex_templates/preamble.tex", "r") as file:
-    tex_preamble = file.read()
-
-with open("tex_templates/end_document.tex", "r") as file:
-    tex_end_document = file.read()
-
 with open("tex_templates/figure.tex", "r") as file:
     tex_figure = file.read()
 
@@ -274,10 +268,9 @@ def cp_images(src_folder, dst_folder):
         latex_section = generate_section(title, text, paper_id, figure_template)
         contributions += "\n\n" + latex_section
 
-full_document = tex_preamble + table + contributions + tex_end_document
-
-with open("3dgs_compression_survey.tex", "w", encoding="UTF-8") as file:
-    file.write(full_document)
+# create contributions
+with open("3dgs_contributions.tex", "w", encoding="UTF-8") as file:
+    file.write(contributions)
 
 # Compile LaTeX Document
 subprocess.run(["pdflatex", "3dgs_compression_survey.tex"])

data_extraction/latex/tex_templates/contributions.tex

@@ -1,5 +1,3 @@
-\section*{Contribution Summaries}
-
 In the following sections, we present a comprehensive overview of state-of-the-art 3D Gaussian Splatting (3DGS) compression methods. 
 Although each method shares the common goal of minimizing memory usage while preserving rendering quality and speed, their strategies differ significantly. 
 The summaries aim to distill the key concepts behind each approach, enabling better comparison and serving as a valuable reference for future research in the field of 3DGS compression.

data_extraction/latex/tex_templates/entry.tex

@@ -1,4 +1,4 @@
-\subsection*{<paper_title>}
+\subsection{<paper_title>}
 
 <paper_text> \cite{<paper_id>}