AI Paper Atlas

Research map / Reinforcement learning / MDPs and planning

Tracking as Online Decision-Making: Learning a Policy From Streaming Videos With Reinforcement Learning

James Supancic,III, Deva Ramanan

ICCV 2017 · Reinforcement learning › MDPs and planning · approach: POMDPs and confounded RL · 117 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
replace online heuristics with a learned policy (offline heuristics plus Q-learning)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .09 -> .30Improved
replace online heuristics with offline-heuristic classifier using track heuristic only (ablation)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .09 -> .13Improved
use offline update heuristic only (ablation)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .09 -> .14Improved
use both offline track and update heuristics (ablation)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .09 -> .20Improved
add Q-learning refinement on top of offline heuristics (ablation)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .20 -> .30Improved
learn only the appearance update action with Q-learning (ablation)p-track (FCNT) · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .28Improved
use CCOT TRACK/UPDATE modules instead of FCNT (ablation)p-track · Long-term Internet videos (16 held-out, >=5,000 frames)F1: .30 -> .36Improved
initialize Q-functions from offline heuristics with supervised regularizer (ablation)deep Q-learning tracker · Long-term Internet videos (16 held-out, >=5,000 frames)convergenceImproved
train a dedicated short-term policy on dissimilar short-term videos (p-track-short) instead of long-term policy (ablation)p-track · OTB-2013 / VOT-2016accuracyImproved
train policy on long-term Internet videos and test on short-term benchmarks (ablation)p-track-long · OTB-2013 / VOT-2016accuracyMixed
train on more videos beyond 8,000-12,000 (ablation)Q-learning tracker · Long-term Internet videos (16 held-out, >=5,000 frames)F1No clear effect
replace bounding-box labels with binary correct/incorrect feedbackannotation · annotation throughputframes per minute: 34 -> 1200Improved

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