AI Paper Atlas

Research map / Video / Video segmentation and tracking

Tracking Persons-of-Interest via Unsupervised Representation Adaptation

Shun Zhang, Jia‐Bin Huang, Jongwoo Lim, Yihong Gong, Jinjun Wang, Narendra Ahuja, Ming–Hsuan Yang

IJCV 2019 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 31 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 SymTriplet loss instead of triplet lossOurs-Triplet · BBT05 clusteringweighted purity: 0.68 -> 0.85Improved
use SymTriplet loss instead of contrastive (Siamese) lossOurs-Siamese · BBT05 clusteringweighted purity: 0.70 -> 0.85Improved
add contextual constraints to spatio-temporal constraints (ablation)Ours-SymTriplet · BBT (multi-face tracking, overall)MOTA (%): 57.2% -> 59.6%Improved
add contextual constraints to spatio-temporal constraints (ablation)Ours-SymTriplet · BBT05 clusteringweighted purity: 0.85 -> 0.86Improved
add linking of tracklets within each shot (ablation)Ours-SymTriplet-Contx · BBT (multi-face tracking, overall)MOTA (%): 43.1% -> 59.6%Improved
add clustering of shot-level tracklets across shots (ablation)Ours-SymTriplet-Contx · BBT (multi-face tracking, overall)MOTA (%): 46.2% -> 59.6%Improved
add clustering across shots (fewer ID switches) (ablation)Ours-SymTriplet-Contx · BBT (multi-face tracking, overall)ID switches (IDS): 1496 -> 817Improved
adapt the CNN features to the specific video instead of pre-trained featurespre-trained face features · BBT (multi-face tracking, overall)MOTA (%): 30.0% -> 57.2%Improved
nonlinear metric learning by fine-tuning all layers instead of Mahalanobis mapping on VGG-FaceVGG-Face-ULDML · T-ara clusteringweighted purity: 0.26 -> 0.84Improved
use two-threshold and Hungarian linking instead of mTLD for shot-level trajectoriesOurs-mTLD · BBT (multi-face tracking, overall)MOTA (%): 44.6% -> 57.2%Improved
train features on one episode (BBT02) and transfer to others instead of per-video adaptationOurs-SymTriplet · BBT05 clusteringweighted purity: 0.85 -> 0.79Hurt
use large feature dimensions (512, 1024) instead of 64 (ablation)64-d features · Bruno Mars clusteringweighted purityHurt

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