AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Reinforcement Learning Meets Visual Odometry

Nico Messikommer, Giovanni Cioffi, Mathias Gehrig, Davide Scaramuzza

ECCV 2024 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 17 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add an RL agent for keyframe and grid-size selection to DSODSO · EuRoCATE avg (m): 0.594 -> 0.483Improved
add an RL agent to SVOSVO · EuRoC V103tracking success: SVO fails -> 0.394Improved
add an RL agent to SVOSVO · TUM-RGBDATE avg (m): 0.471 -> 0.422Improved
add an RL agent to DSO on TUM-RGBD (rolling shutter)DSO · TUM-RGBDsequences completed: two more sequences completed; some per-sequence errors higherMixed
add an RL agent to ORB-SLAM3ORB-SLAM3 · EuRoCATE avg (m): 0.573 -> 0.475Improved
add keyframe selection action to the agent (ablation)RL SVO · EuRoCrobustnessImproved
add grid size selection action (ablation)RL SVO · EuRoCATE avg (m): 0.799 -> 0.696Improved
add keypoint inputs via Variable Encoder (ablation)RL SVO · EuRoCsequences completedImproved
train with privileged critic (ablation)RL SVO · EuRoCATE avg (m): 0.754 -> 0.696Improved
use RL agent for keyframe selectionSVO · EuRoCruntime (ms): 9.06 -> 7.40 (SVO); 34.45 -> 32.29 (DSO)Improved
RL-tuned classical VO versus learned hybrid VO (DPVO)RL DSO / RL SVO · EuRoCATE avg (m): DPVO 0.105 vs RL DSO 0.320 vs RL SVO 0.972Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy