AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Self-Supervised Deep Visual Odometry With Online Adaptation

Shunkai Li, Xin Wang, Yingdian Cao, Fei Xue, Zike Yan, Hongbin Zha

CVPR 2020 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 79 citations (OpenAlex)

Check your idea against 137,217 papers

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
use next-window loss after an inner-loop update (meta-learning) instead of naive online learning (ablation)Self-supervised VO · KITTI Seq. 09t_err (%): 10.22 -> 5.89Improved
use next-window loss after an inner-loop update (meta-learning) instead of naive online learning (ablation)Self-supervised VO · KITTI Seq. 10t_err (%): 8.24 -> 4.79Improved
add convLSTM to meta-learning (ablation)Self-supervised VO · KITTI Seq. 09t_err (%): 9.25 -> 6.36Improved
add convLSTM to meta-learning (ablation)Self-supervised VO · KITTI Seq. 10t_err (%): 7.58 -> 5.37Improved
add feature alignment to meta-learning (ablation)Self-supervised VO · KITTI Seq. 09t_err (%): 9.25 -> 7.52Improved
add feature alignment to meta-learning (ablation)Self-supervised VO · KITTI Seq. 10t_err (%): 7.58 -> 5.98Improved
add convLSTM to meta-learning with feature alignment (ablation)Self-supervised VO · KITTI Seq. 09t_err (%): 7.52 -> 5.89Improved
add convLSTM to meta-learning with feature alignment (ablation)Self-supervised VO · KITTI Seq. 10t_err (%): 5.98 -> 4.79Improved
add feature alignment to meta-learning with convLSTM (ablation)Self-supervised VO · KITTI Seq. 09t_err (%): 6.36 -> 5.89Improved
add feature alignment to meta-learning with convLSTM (ablation)Self-supervised VO · KITTI Seq. 10t_err (%): 5.37 -> 4.79Improved
replace meta-learning with naive online learning (ablation)Self-supervised VO with LSTM and feature alignment · KITTI Seq. 09r_err (°/100m): 3.34 -> 5.33Hurt
increase sliding-window length up to 15 frames (ablation)Self-supervised VO · KITTIadaptation speed and accuracyImproved

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