The objective of this project is to acquire a grasp of fundamental concepts and techniques for reconstructing a real-world scene from multiple images, even when not previously calibrated. The primary focus is on the three-dimensional recovery of urban environments, utilizing diverse datasets that include images of facades and aerial views of cities.
This endeavour holds practical significance across various applications where deriving three-dimensional information from images captured from varied perspectives is essential. Examples of such applications encompass image mosaics or panoramas, augmented reality, depth computation, 3D reconstruction, 3D localization and navigation, as well as the synthesis of new viewpoints.
The results of each of the lab sessions can be found in the folder Reports
# clone project
git clone https://github.com/aolivtous/MCV_M4_3DVision.git
cd MCV_M4_3DVision
# [OPTIONAL] create conda environment
conda create -n m4 python=3.10
conda activate m4
# install requirements
conda install --file requirements.txt
# OR
pip install -r requirements.txt