Skip to content
RobothecaBeta
Log in

DROID-SLAM

  • Software
  • Navigation and mapping
  • Library

A deep learning visual SLAM system that estimates camera poses and builds maps from monocular, stereo, or RGB-D camera input. It is built for research and evaluation on standard datasets.

  • Stale

Made by Princeton Vision & Learning Lab, United States (checked on github.com/princeton-vl (opens in a new tab)).

Source on GitHub: github.com/princeton-vl/DROID-SLAM (opens in a new tab). Facts read at commit 2dfd39f (opens in a new tab).

Licenses

Code
BSD-3-Clause (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (BSD-3-Clause) at commit 2dfd39f (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

Labels

  • StaleNo commit in 12 months: the newest is from 5 May 2025.

What it does

  • Performs visual SLAM using a learned model, accepting monocular, stereo, or RGB-D cameras.
  • Runs demos on video image folders using a calibration file that gives the camera intrinsics.
  • Can save a reconstruction with full-resolution depth maps and view it with a provided script.
  • Offers an asynchronous mode that runs frontend and backend in separate processes, optionally on multiple GPUs.
  • Includes evaluation scripts for TartanAir, EuRoC, TUM-RGBD, and ETH3D-SLAM datasets.
  • Provides training code, trained on the TartanAir dataset.

Written by Robotheca from the project's README (opens in a new tab) at commit 2dfd39f.

Getting started

Difficulty
Advanced
Hardware it needs, as the project states it
EuRoC and TUM evaluations can run on a 1080Ti; TartanAir and ETH3D evaluations need 24G of memory.

Maintenance

Status
Stale: no commit in 12 months
Newest commit
5 May 2025 on main
Latest release
None published on GitHub

Checked on GitHub on 16 Oct 2026.

Links

Robots that can run it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.