AI Paper Atlas

Research map / Detection and segmentation / Tracking

PathTrack: Fast Trajectory Annotation With Path Supervision

Santiago Manen, Michael Gygli, Dengxin Dai, Luc Van Gool

ICCV 2017 · Detection and segmentation › Tracking · approach: Association cues for online MOT · 69 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
train the LP tracker's person matcher on PathTrack instead of MOT15LP tracker · PathTrack test setMOTA: 24.5 -> 27.6Improved
train the LP tracker's person matcher on PathTrack instead of MOT15LP tracker · PathTrack test setID Switches: 1,827 -> 1,576Improved
add PathTrack-trained person matcher (TRID) to relink NOMTwSDP trajectoriesNOMTwSDP · MOT15 testID Switches: 427 -> 351Improved
add PathTrack-trained person matcher (TRID) to relink NOMTwSDP trajectoriesNOMTwSDP · MOT15 testMOTA: 55.5 -> 55.7Improved
add PathTrack-trained person matcher (TRID) to relink NOMTwSDP trajectoriesNOMTwSDP · MOT15 testFP: 5,594 -> 6,273Hurt
train person matching on PathTrack (15,380 trajectories) instead of MOT15 (521)Siamese CNN · PathTrack test setmatching accuracy: 78% -> 88%Improved
train with PathTrack when context features are usedSiamese CNN + context features · PathTrack test setmatching accuracy: 84% -> 90%Improved
use high-quantity strategy (140 PathTrack videos) instead of high-quality (11 MOT15 videos) at equal annotation time (ablation)Siamese CNN · PathTrack test setmatching accuracy: 78% -> 85%Improved
increase number of training trajectories toward 15,380 (ablation)Siamese CNN · PathTrack test setmatching accuracyImproved
add path supervision to box annotationsLabelMe/VATIC box annotation · MOT15 train annotation accuracyrecall at IoUImproved
add path supervision versus box-only annotation, large budgetsVATIC / LabelMe · MOT15 train annotation accuracyrecall at IoUNo clear effect
use path supervision instead of box annotation for the same accuracyVATIC / LabelMe · MOT15 trainannotation time: almost twice and three times more timeImproved

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