AI Paper Atlas

Research map / Video / Video segmentation and tracking

Video Object Segmentation with Episodic Graph Memory Networks

Xiankai Lu, Wenguan Wang, Martin Danelljan, Tianfei Zhou, Jianbing Shen, Luc Van Gool

ECCV 2020 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 292 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 graph memory network (ablation)backbone · DAVIS17 (one-shot)mean J: 73.5 -> 80.0Improved
add graph memory network (ablation)backbone · DAVIS17 (one-shot)mean F: 78.4 -> 85.9Improved
add graph memory network (ablation)backbone · DAVIS16 (zero-shot)mean J: 73.2 -> 82.5Improved
add graph memory network (ablation)backbone · DAVIS16 (zero-shot)mean F: 72.5 -> 81.2Improved
increase memory from 2 nodes to 3 nodes (ablation)graph memory · DAVIS17 (one-shot)mean J: 76.0 -> 80.0Improved
increase memory from 3 to 5 nodes (ablation)graph memory · DAVIS17 (one-shot)mean J: 80.0 -> 80.0No clear effect
increase memory from 3 to 4 nodes (ablation)graph memory · DAVIS16 (zero-shot)mean J: 82.5 -> 82.5No clear effect
add iterative memory updating (K=0 to K=4) (ablation)graph memory · DAVIS17 (one-shot)mean J: 78.1 -> 80.0Improved
add iterative memory updating (K=0 to K=4) (ablation)graph memory · DAVIS17 (one-shot)mean F: 82.2 -> 85.9Improved
add iterative memory updating (K=0 to K=4) (ablation)graph memory · DAVIS16 (zero-shot)mean J: 81.2 -> 82.5Improved
increase update steps from K=2 to K=4 (ablation)graph memory · DAVIS17 (one-shot)mean J: 79.3 -> 80.0Improved
add label shuffling during training (ablation)graph memory · DAVIS17 (one-shot)mean J: 78.5 -> 80.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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