AI Paper Atlas

Research map / Detection and segmentation / Tracking

Real-Time Object Tracking via Meta-Learning: Efficient Model Adaptation and One-Shot Channel Pruning

Ilchae Jung, Kihyun You, Hyeonwoo Noh, Minsu Cho, Bohyung Han

AAAI 2020 · Detection and segmentation › Tracking · approach: Foundation-model and correlation-filter tracking · 35 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
Use meta-learned adaptation instead of standard RT-MDNet adaptationRT-MDNet · OTB2015success (%): 65.0 -> 65.5Improved
Use meta-learned adaptation with 5 initial and 5 online iterationsRT-MDNet · OTB2015FPS: 42 -> 58Improved
Add estimated-target online adaptation (ablation)MetaRTT · OTB2015success (%): 58.6 -> 65.5Improved
Add hard examples to meta-training (ablation)MetaRTT · OTB2015success (%): 60.7 -> 65.5Improved
Use per-parameter and per-iteration learning rates instead of a scalar rate (ablation)MetaRTT · OTB2015success (%): 60.9 -> 65.5Improved
Add one-shot channel pruningMetaRTT · OTB2015FPS: 58 -> 65Improved
Add one-shot channel pruningMetaRTT · OTB2015success (%): 65.5 -> 65.1Hurt
Meta-learn the pruning network over complete simulated episodes (ablation)MetaRTT+Prune · OTB2015success (%): 61.7 -> 65.1Improved
Use input-adaptive pruning instead of a universal static mask (ablation)MetaRTT+Prune · OTB2015success (%): 60.9 -> 65.1Improved
Reduce RT-MDNet adaptation to 5 initial and 5 online iterations without meta-learning (ablation)RT-MDNet · OTB2015success (%): 65.0 -> 57.8Hurt
Add COCO pretrainingMetaRTT · VOT2016EAO: 0.346 -> 0.350Improved

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