AI Paper Atlas

Research map / Video / Video segmentation and tracking

Video Instance Segmentation

Linjie Yang, Yuchen Fan, Ning Xu

ICCV 2019 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 541 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 category consistency cue to the matching score (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 24.7 -> 30.3Improved
add box IoU cue to the matching score (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 29.2 -> 30.3Improved
add detection confidence cue to the matching score (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 29.5 -> 30.3Improved
add all three cues over appearance-only matching (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 21.1 -> 30.3Improved
add category cue alone to appearance-only matching (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 21.1 -> 27.9Improved
add box IoU cue alone to appearance-only matching (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 21.1 -> 22.7Improved
add detection score alone to appearance-only matching (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 21.1 -> 23.4Improved
replace predicted identities with ground truth identities (Identity Oracle) (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 30.3 -> 31.5No clear effect
replace predicted image-level detections with ground truth (Image Oracle) (ablation)MaskTrack R-CNN · YouTube-VIS validationAP: 30.3 -> 78.7Improved
replace tracking-by-detection baseline SeqTracker with end-to-end tracking branchMaskTrack R-CNN · YouTube-VIS testAP: 29.5 -> 32.3Improved
replace IoUTracker+ with learned appearance tracking branchMaskTrack R-CNN · YouTube-VIS validationAP: 23.6 -> 30.3Improved
replace mask propagation (FEELVOS) with tracking branchMaskTrack R-CNN · YouTube-VIS validationAP: 26.9 -> 30.3Improved

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