ChenHoy/DROID-Splat
An end-to-end SLAM framework combining DROID-SLAM, monocular depth priors, and 3D Gaussian Splatting for real-time dense 3D reconstruction from video.
Not currently ranked — collecting fresh signals.
star history
This project builds a deep-learning-based dense visual SLAM system that performs real-time global optimization of camera poses and 3D geometry. It combines state-of-the-art tracking from DROID-SLAM, monocular depth estimation priors (Metric3D), and differentiable rendering via 3D Gaussian Splatting. The system supports camera intrinsics optimization through DroidCalib and offers alternative branches using MCMC-based densification and 2D Gaussian Splatting for faster convergence.
Frequently asked
- What is ChenHoy/DROID-Splat?
- An end-to-end SLAM framework combining DROID-SLAM, monocular depth priors, and 3D Gaussian Splatting for real-time dense 3D reconstruction from video.
- Is DROID-Splat open source?
- Yes — ChenHoy/DROID-Splat is open source, released under the Apache-2.0 license.
- What language is DROID-Splat written in?
- ChenHoy/DROID-Splat is primarily written in Python.
- How popular is DROID-Splat?
- ChenHoy/DROID-Splat has 414 stars on GitHub.
- Where can I find DROID-Splat?
- ChenHoy/DROID-Splat is on GitHub at https://github.com/ChenHoy/DROID-Splat.