Judging the Relevance and worth of ideas part 2.pptx
B Srinivas(1RV19IS014)_Internship_Phase_2 (1).pptx
1. Department of Information Science and Engineering
Internship(18IS74)
7th Semester BE
3D RECONSTRUCTION OF HISTORICAL MONUMENTS
Evaluation- Phase-2
Academic Year – 2022-23
Carried out by:
B SRINIVAS (1RV19IS014)
Carried out at:
Bhargawa Info Tech Solutions (P) Ltd.
External Guide and Designation
Dr. Anala M.R
Professor, Dept Of ISE
R V College of Engineering
2. Agenda
• Objectives of the project
• Tasks performed
• Tools/Techniques Used
• Results
• Observation and conclusion
• References
4. This Project will focus on direct reconstruction of the historical
monuments using Blender, a open-source 3d render using images
and GIS as the reference.
Using images and information available the monuments will be
covered in textures for a realistic and clear understanding of the
monument.
The Project will also utilize Unity, a game development engine to
develop a virtual world for the virtual inspection of the monument.
RV College of
Engineering
Objectives
Go, change the world
5. This Project will focus on 3D reconstruction of the historical
monuments using Blender, a open-source 3d render using images
and Images as the reference.
Using images and information available the monuments will be
covered in textures for a realistic and clear understnding of the
monument.
The Models will be reconstructed using the images using pixel
extraction from the images and Mesh conversion techniques.
RV College of
Engineering
Objectives
Go, change the world
6. • VSCode.
• Blender.
• Unity.
• Adobe Photoshop.
• Python.
• Pixel Extraction and Mesh Conversion.
RV College of
Engineering
Tools and Techniques Used
Go, change the world
7. Design and Implementation
• The algorithm uses images from a pre-specified directory and
runs masking applications such as GAN.
• GAN Networks are used so as to distinguish the required
image from the background image.
• The output of the GAN Networks is used to separate the
foreground of the image from the background.
• The extracted Image is then imported into the Blender.
• Blender imports these images as planes with each of the pixel
as a vertex.
• The RGB values of the pixel is used to determine the color of
the vertex.
8. Design and Implementation
• The algorithm then performs certain optimizations to reduce
the number of pixels for easier rendering.
• The algorithm then performs the above mentioned steps for
multiple such images and stores the meshes.
• The meshes are then applied with the solidification modifier
and the meshes are thickened
• The meshes are then aligned together by performing regular
checks with the mesh collision and overlapping and is brute-
forced along the axises until they form a perfect structure.
• The excess of the mesh is then trimmed using the Boolean
modifiers based on their overlaps and the solidification
modifier is removed to obtain the final mesh.
9. [1]. Lemonia Ragia , Froso Sarri and Katerina Mania, ‘3D Reconstruction and
Visualization of Alternatives for Restoration of Historic Buildings A New Approach’,
GITSAM, 2015.
[2]. Sabry F. El-Hakim, J.-Angelo Beraldin, Michel Picard, and Guy Godin, ‘Detailed 3D
reconstruction of large-scale heritage sites with integrated techniques’, IEEE, 2004.
[3]. Antonios Kargas, Georgios Loumos and Dimitrios Varoutas, ‘Using Different Ways
of 3D Reconstruction of Historical Cities for Gaming Purposes: The Case Study of
Nafplio’ , Heritage, 2019.
[4]. Dagmāra Krūmiņa, ‘Cultural heritage 3D reconstruction from historical materials’ ,
Riga Technical University, Riga, Latvia, Student Contest, 2019.
[5]. Ioannis Liritzis, Pantelis Volonakis and Spyros Vosinakis, ‘3D Reconstruction of
Cultural Heritage Sites as an Educational Approach. The Sanctuary of Delphi’, applsci,
2021
RV College of
Engineering
References
Go, change the world