AI Paper Atlas

Research map / Detection and segmentation / Tracking

Homography Decomposition Networks for Planar Object Tracking

Xinrui Zhan, Yueran Liu, Jianke Zhu, Yang Li

AAAI 2022 · Detection and segmentation › Tracking · approach: Siamese network tracking · 21 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
decompose homography estimation into similarity and residual stages (ablation)HDN · POT-210Prec: 0.319 -> 0.841Improved
add the residual component to similarity-only tracking (ablation)HDN · POT-210Prec: 0.689 -> 0.841Improved
combine supervised and unsupervised residual losses (ablation)HDN · POT-210Prec: 0.837 -> 0.841; 0.417 -> 0.841Improved
add classification and regression losses (ablation)HDN · POT-210Prec: 0.732 -> 0.841Improved
add rotated classification labels (ablation)HDN · POT-210Prec: 0.836 -> 0.841Improved
add residual negative-sample loss (ablation)HDN · POT-210Prec: 0.733 -> 0.841Improved
use HDN instead of LISRDplanar tracking · POT-210avg Prec: 0.752 -> 0.841Improved
use HDN instead of GOP-ESMplanar tracking · POICPrec (e≤5): 0.868 -> 0.749Hurt

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