AI Paper Atlas

Research map / Detection and segmentation / Tracking

Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackers

Eunbyung Park, Alexander C. Berg

ECCV 2018 · Detection and segmentation › Tracking · approach: Siamese network tracking · 191 citations (OpenAlex)

Read the paperCheck 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
add meta-learned initialization (1 iteration) to CRESTCREST · VOT2016 (CREST family)EAO: 0.283 -> 0.317Improved
add meta-learned initialization (1 iteration) to CRESTCREST-01 · VOT2016 (CREST family)EAO: 0.259 -> 0.317Improved
add meta-learned initialization to CRESTCREST · VOT2016 (CREST family)Robustness (failures): 1.083 -> 0.932Improved
add meta-learned initialization (1 iteration) to pyMDNetpyMDNet-30 · VOT2016 (MDNet family)EAO: 0.304 -> 0.314Improved
add meta-learned initialization to pyMDNet at 1 iterationpyMDNet-01 · VOT2016 (MDNet family)EAO: 0.119 -> 0.314Improved
add meta-learned initialization to pyMDNetpyMDNet-30 · VOT2016 (MDNet family)Accuracy: 0.540 -> 0.526Hurt
reduce initial-frame iterations from 30 to 1 in pyMDNet (no meta-learning) (ablation)pyMDNet · VOT2016 (MDNet family)EAO: 0.304 -> 0.119Hurt
reduce initial-frame iterations from 10 to 1 in CREST (no meta-learning) (ablation)CREST · VOT2016 (CREST family)EAO: 0.252 -> 0.259No clear effect
add meta-learning to cut initial-frame training time in MDNetpyMDNet-30 · VOT2016 (MDNet family)training time (s): 3.508 -> 0.124Improved
add meta-learning to cut initial-frame training time in CRESTCREST-01 · VOT2016 (CREST family)training time (s): 0.395 -> 0.090Improved
add meta-learned initialization to improve generalization to future framesCREST-65 · VOT2016 (CREST family)lookahead loss: 0.30 -> 0.05Improved
increase meta-update iterations T beyond 1 (ablation)MetaSDNet · meta-trainingtracking performanceNo clear effect

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