AI Paper Atlas

Research map / Video / Video segmentation and tracking

Anchor Diffusion for Unsupervised Video Object Segmentation

Zhao Yang, Qiang Wang, Luca Bertinetto, Weiming Hu, Song Bai, Philip H. S. Torr

ICCV 2019 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 119 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 intra-frame non-local branch (ablation)DeepLabv3 baseline · DAVIS-2016 valregion similarity J: +0.76Improved
add intra-frame non-local branch (ablation)DeepLabv3 baseline · DAVIS-2016 valcontour accuracy F: -0.20No clear effect
concatenate anchor frame features (ablation)DeepLabv3 baseline · DAVIS-2016 valregion similarity J: +1.43Improved
concatenate anchor frame features (ablation)DeepLabv3 baseline · DAVIS-2016 valcontour accuracy F: +0.18No clear effect
add anchor diffusion (ablation)DeepLabv3 baseline · DAVIS-2016 valregion similarity J: +2.02Improved
add anchor diffusion (ablation)DeepLabv3 baseline · DAVIS-2016 valcontour accuracy F: +1.20Improved
add intra-frame branch on top of anchor diffusion (full AD-Net, single scale) (ablation)Baseline + anchor-diffusion · DAVIS-2016 valregion similarity J: 77.43 -> 78.26Improved
add intra-frame branch on top of anchor diffusion (full AD-Net, single scale) (ablation)Baseline + anchor-diffusion · DAVIS-2016 valcontour accuracy F: 76.78 -> 77.11Improved
use multi-scale and mirrored inputs at inferenceAD-Net · DAVIS-2016 valregion similarity J: 78.26 -> 79.4Improved
add instance pruning post-processingAD-Net · DAVIS-2016 valregion similarity J: 79.4 -> 81.7Improved
add instance pruning post-processingAD-Net · DAVIS-2016 valcontour accuracy F: 78.2 -> 80.5Improved
sample both training frames randomly instead of using first frame as anchor (ablation)AD-Net · DAVIS-2016 valsegmentation qualityHurt

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