AI Paper Atlas

Research map / Video / Video segmentation and tracking

Interactive Video Object Segmentation Using Global and Local Transfer Modules

Yuk Heo, Yeong Jun Koh, Chang-Su Kim

ECCV 2020 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 33 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 local transfer module (ablation)Proposed interactive VOS · DAVIS2017 val, round 1J: 0.629 -> 0.676Improved
add local transfer module (ablation)Proposed interactive VOS · DAVIS2017 val, round 5J: 0.771 -> 0.783Improved
use front TE instead of rear TE features for local affinity (ablation)Proposed interactive VOS · DAVIS2017 val, round 1J: 0.653 -> 0.676Improved
use front TE instead of rear TE features (ablation)Proposed interactive VOS · DAVIS2017 val, round 5J: 0.760 -> 0.783Improved
add auxiliary loss (lambda 0 -> 0.1) (ablation)Proposed interactive VOS · DAVIS2017 val, round 1J: 0.645 -> 0.676Improved
add auxiliary loss (lambda 0 -> 0.1) (ablation)Proposed interactive VOS · DAVIS2017 val, round 5J: 0.761 -> 0.783Improved
lambda 0.5 instead of 0.1 (ablation)Proposed interactive VOS · DAVIS2017 val, round 1J: 0.676 -> 0.658Hurt
lambda 1.0 instead of 0.1 (ablation)Proposed interactive VOS · DAVIS2017 val, round 5J: 0.783 -> 0.762Hurt
use probability transfer modules instead of matching (A-Net predictions) (ablation)Proposed interactive VOS · DAVIS2017 valAUC-J: 0.636 -> 0.771Improved
use probability transfer modules instead of matching (scribbles) (ablation)Proposed interactive VOS · DAVIS2017 valJ&F@60s: 0.690 -> 0.827Improved
compare with Oh et al. in user studyProposed interactive VOS · DAVIS2016 user studyseconds per video: 37.9 -> 29.8Improved
compare with Oh et al. in user studyProposed interactive VOS · DAVIS2016 user studyrounds per video: 2.77 -> 1.90Improved

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