voyager / data_engine /README.md
ezeanubis's picture
Upload folder using huggingface_hub
a7aea10 verified

A newer version of the Gradio SDK is available: 6.1.0

Upgrade

This is the data engine for processing training data of Voyager.

πŸ› οΈ Dependencies and Installation

Begin by cloning required repositories:

# VGGT
git clone https://github.com/facebookresearch/vggt.git
touch vggt/vggt/__init__.py  # create an empty init.py

# MoGe
git clone https://github.com/microsoft/MoGe.git

# Metric3D
git clone https://github.com/YvanYin/Metric3D.git
# comment out line 8-12 in Metric3D/mono/utils/comm.py
# add from mono.model.backbones import * to Metric3D/mono/utils/comm.py

Install required dependencies:

conda create -n data_engine python=3.10
conda activate data_engine
pip install -r requirements.txt

πŸ› οΈ Install Environment

# project path
cd data_engine

# VGGT
git clone https://github.com/facebookresearch/vggt.git
touch vggt/vggt/__init__.py

# MoGe
git clone https://github.com/microsoft/MoGe.git

# Metric3D
git clone https://github.com/YvanYin/Metric3D.git
# !!! important steps:
# comment out line 8-12 in Metric3D/mono/utils/comm.py
# and then add from mono.model.backbones import * to Metric3D/mono/utils/comm.py

# pip install environment
conda create -n voyager_dataengine python=3.10
conda activate voyager_dataengine
pip install -r requirements.txt

# run dataEngine
bash dataEngine.sh

πŸ”‘ Run Data Engine

We provide a script to run the data engine.

bash run.sh