atlas

Research map / Detection and segmentation / Tracking

A Twofold Siamese Network for Real-Time Object Tracking

CVPR 2018 · Detection and segmentation › Tracking · approach: Siamese network tracking · 666 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
combine semantic and appearance branches (ablation)SiamFC · OTB-2013AUC: 0.599 / 0.607 -> 0.649Helped
add multilevel semantic features (ablation)SA-Siam · OTB-2013AUC: 0.649 -> 0.656Helped
add multilevel semantic features (ablation)SA-Siam · OTB-50AUC: 0.583 -> 0.581Hurt
add channel attention (ablation)SA-Siam · OTB-50AUC: 0.583 -> 0.591Helped
add channel attention (ablation)SA-Siam · OTB-100AUC: 0.637 -> 0.642Helped
use heterogeneous semantic and appearance models rather than ensembles of similar models (ablation)SA-Siam · OTB-2013AUC: A1A2: 0.608; S1S2: 0.602; SA-Siam: 0.649Helped
jointly train the two branches instead of training separately (ablation)SA-Siam · OTB-2013/50/100AUC: 0.630, 0.546, 0.620Hurt
use three-scale searchSA-Siam · object trackingspeedMixed
fine-tune S-NetSA-Siam · object trackingtracking performanceHurt

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