AI Paper Atlas

Research map / Detection and segmentation / Tracking

"HVC-Net: Unifying Homography, Visibility, and Confidence Learning for Planar Object Tracking"

Haoxian Zhang, Yonggen Ling

ECCV 2022 · Detection and segmentation › Tracking · approach: Foundation-model and correlation-filter tracking · 12 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
add correlation block (D) (ablation)Ours w/o DPMR · Synthetic MS-COCO test setAE: 6.678 -> 4.173Improved
add pyramids (P) to correlation block model (ablation)Ours-D · Synthetic MS-COCO test setAE: 4.173 -> 1.145Improved
add pyramids (P) without correlation block (ablation)Ours w/o DPMR · Synthetic MS-COCO test setAE: 6.678 -> 5.280Improved
add refinement module (R) (ablation)Ours-DP · Synthetic MS-COCO test setAE: 1.145 -> 0.876Improved
add refinement module (R) (ablation)Ours-P · Synthetic MS-COCO test setAE: 5.280 -> 2.970Improved
add joint visibility learning (M) (ablation)Ours-DP · Synthetic MS-COCO test setAE: 1.145 -> 1.097Improved
add joint visibility learning (M) (ablation)Ours-P · Synthetic MS-COCO test setAE: 5.280 -> 4.051Improved
add joint visibility learning (M) to refined model (ablation)Ours-DPR · Synthetic MS-COCO test setHD: 1.739 -> 1.695Improved
add refinement module (R) (ablation)Ours-DPM · Synthetic MS-COCO test setHD: 2.107 -> 1.695Improved
add correlation block (D) for visibility (ablation)Ours-PMR · Synthetic test set, visibilityvisibility loss: 0.346 -> 0.328Improved
learn visibility jointly (M) rather than independently (V) (ablation)Ours-DPVR · Synthetic test set, visibilityvisibility loss: 0.335 -> 0.328Improved
use OursCon (cost-volume confidence) instead of PatchConOurs-DPMR · Confidence classificationFPR: 14.5% -> 8.7%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