Co-SLAM-Visualizer

Visualization tool similar to NICE-SLAM for Co-SLAM.

Installation

Same as Neural SLAM Evaluation Benchmark

This repo assumes you already configured the environment from Co-SLAM main repository. You then also need the following dependencies:

Open3D
pyglet
pyrender

You can install those dependencies by running:

conda activate coslam
pip install -r requirements.txt

Run Visualizer

Only tested on Rellica, TUM, Azure datasets.

Recommend Step

git clone https://github.com/Immortalqx/Co-SLAM-Visualizer.git
cd Co-SLAM-Visualizer

Copy configs, tools, visualizer.py to the Co-SLAM directory;
Co-SLAM saves the mesh every 500 rounds. If you want to save the mesh every 50 rounds like NICE-SLAM, you need to re-run Co-SLAM.

You can run Co-SLAM Visualizer using the code below:

python visualizer.py --config './configs/{Dataset}/{scene}.yaml 

# Example
python visualizer.py --config ./configs/Replica/room0.yaml
python visualizer.py --config ./configs/Azure/apartment.yaml
python visualizer.py --config ./configs/Tum/fr1_desk.yaml