AI Paper Atlas

Research map / Detection and segmentation / Tracking

Tracklet Association by Online Target-Specific Metric Learning and Coherent Dynamics Estimation

Bing Wang, Gang Wang, Kap Luk Chan, Li Wang

TPAMI 2016 · Detection and segmentation › Tracking · approach: Memory and correlation-based MOT · 83 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 common class metric with target-specific metric learning (TSML) (ablation)tracklet association · PETS2009-S2L1MOTA: 92.1% -> 93.4%Improved
replace common class metric with target-specific metric learning (TSML) (ablation)tracklet association · TUD CrossingIDS / correctly matched: 10/845 -> 7/862Improved
replace common class metric with target-specific metric learning (TSML) (ablation)tracklet association · ETHID switches: 19 -> 5Improved
add tracklet dynamics (TD) to TSML (ablation)tracklet affinity model · PETS2009-S2L1MOTA: 93.4% -> 94.7%Improved
add tracklet dynamics (TD) to TSML (ablation)tracklet affinity model · Town CentreMOTA: 57.3% -> 61.6%Improved
add tracklet dynamics (TD) to TSML (ablation)tracklet affinity model · TUD-StadtmitteMOTA: 97.3% -> 97.3%No clear effect
add learned weighting parameters (WP) (ablation)TSML+TD · PETS2009-S2L1MOTA: 94.7% -> 95.3%Improved
add learned weighting parameters (WP) (ablation)TSML+TD · Town CentreMOTA: 61.6% -> 66.8%Improved
add learned weighting parameters (WP) (ablation)TSML+TD · TUD-StadtmitteMOTA: 97.3% -> 97.3%No clear effect
add one round of tracklet refinement (ablation)tracking algorithm · PETS, TUD, Town Centre, ETHMOTAImproved
use DPM detections instead of public detectionsfull model · MOTChallenge 2DMOTA: 34.3% -> 49.1%Improved
use DPM detections instead of public detectionsfull model · PETS2009-S2L2MOTA: 51.5% -> 59.7%Improved

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