AI Paper Atlas

Research map / Video / Video segmentation and tracking

XMem++: Production-level Video Segmentation From Few Annotated Frames

Maksym Bekuzarov, Ariana Bermudez, Joon-Young Lee, Hao Li

ICCV 2023 · Video › Video segmentation and tracking · approach: Matching-based video object segmentation · 54 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 permanent working memory storing annotated framesXMem · LVOS validation, 10 annotated framesJ: 62.62 -> 75.74Improved
add permanent working memory storing annotated framesXMem · MOSE, 5 annotated framesJ: 67.95 -> 77.11Improved
use XMem++ with 5 annotations instead of competitors with 10XMem++ · LVOS validationJ: 67.36 (5 frames) vs 65.07 best baseline (10 frames)Improved
add permanent memory with one annotationXMem · LVOS and MOSE, 1 frameJNo clear effect
add permanent memory on challenging partial segmentation sequences, with fewer annotationsXMem++ · PUMaVOSframe usage efficiency: 45 -> 10 annotations at similar J (85.43 vs 85.45)Improved
use attention-based frame selection instead of expertsXMem++ · PUMaVOS user studyIoU: 0.805 (experts) vs 0.777 (algorithm) vs 0.779 (non-experts)No clear effect
use attention-based frame selection instead of uniform sampling (ablation)XMem++ · LVOS validationJ and FNo clear effect
disable temporary memory so only the annotated frames remain (ablation)XMem++ · LVOS validationsegmentation qualityImproved
add Gaussian blur augmentation of annotated frames in permanent memory (ablation)XMem++ · LVOS validationsegmentation quality: +0.019Improved
add translate right +100 px augmentation (ablation)XMem++ · LVOS validationsegmentation quality: -0.037Hurt
add grayscale augmentation (ablation)XMem++ · LVOS validationsegmentation quality: -0.005Hurt
add crop-mask-region augmentation (ablation)XMem++ · LVOS validationsegmentation quality: -0.008Hurt

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