Project installation Instructions...
Linux only!
This is a sample hardware setup
Nvidia 3070 GPU
Intel 12700k
32G DDR5 Ram
Smart people focus on the right things..
How to Use
3D Reconstruction Project
This project aims to create an easy-to-use 3D reconstruction pipeline, allowing users to reconstruct scenes from a single input mp4 file taken with any phone or camera. By leveraging powerful AI and deep learning algorithms, the pipeline can reconstruct 3D scenes even from decades-old videos, providing a new way to recreate memories.
As an undergraduate with limited time and resources, there is plenty of room for improvement in the pipeline. Future development will focus on refining the pipeline and overcoming constraints such as inadequate hardware, sophisticated AI systems, and deep learning algorithms.
- Ensure you have all the required libraries installed (see the list of requirements below).
- Open the Jupyter Notebook by running
jupyernotebookand then openingMain_script.ipynb.
Requirements
This project requires specific hardware and software to function properly. AI 3D modeling is extremely difficult to set up a proper training environment, and currently, only Nvidia GPUs are supported. The project has been tested on Ubuntu 20.04 with CUDA 11.1 and 30th Nvidia GPU only.
Install the following system dependencies:
sudo apt-get install \
git \
cmake \
build-essential \
libboost-program-options-dev \
libboost-filesystem-dev \
libboost-graph-dev \
libboost-system-dev \
libboost-test-dev \
libeigen3-dev \
libsuitesparse-dev \
libfreeimage-dev \
libmetis-dev \
libgoogle-glog-dev \
libgflags-dev \
libglew-dev \
qtbase5-dev \
libqt5opengl5-dev \
libcgal-dev
Install Python libraries with pip:
pip install torch==1.10.2+cu111 torchvision==0.11.3+cu111 --extra-index-url https://download.pytorch.org/whl/cu111 --no-cache-dir
pip install torch-scatter -f https://data.pyg.org/whl/torch-1.10.2+cu111.html
pip install -r requirements.txt
pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch
Follow the instructions in the README to install other dependencies, such as Ceres Solver and COLMAP.
