AI Paper Atlas

Research map / Video / Video segmentation and tracking

Segment Every Reference Object in Spatial and Temporal Spaces

Jiannan Wu, Yi Jiang, Bin Yan, Huchuan Lu, Zehuan Yuan, Ping Luo

ICCV 2023 · Video › Video segmentation and tracking · approach: Matching-based video object segmentation · 19 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
multi-task joint training instead of single-task training (ablation)UniRef++-R50 · Ref-Youtube-VOSJ&F: 58.3 -> 60.1Improved
multi-task joint training instead of single-task training (ablation)UniRef++-R50 · RefCOCOoIoU: 72.1 -> 74.7Improved
multi-task joint training instead of single-task training (ablation)UniRef++-R50 · FSS-1000mIoU: 81.8 -> 80.3Hurt
share one UniFusion module across tasks (vs per-level-only) (ablation)UniFusion · FSS-1000mIoU: 75.2 -> 80.3Improved
share one UniFusion module across tasks and levels (vs task-shared only) (ablation)UniFusion · FSS-1000mIoU: 64.1 -> 80.3Improved
share UniFusion across visual levels (ablation)UniFusion · Youtube-VOS2018G: 80.9 -> 81.9Improved
increase object queries from 1 to 100 (ablation)UniRef++-R50 · Youtube-VOS2018G: 79.0 -> 81.9Improved
increase object queries from 100 to 300 (ablation)UniRef++-R50 · RefCOCOoIoU: 75.1 -> 74.7No clear effect
add mask reference (previous-frame mask propagation) for RVOS (ablation)UniRef++-R50 · Ref-Youtube-VOSJ&F: 60.4 -> 61.5Improved
add mask reference for RVOS (ablation)UniRef++-L · Ref-Youtube-VOSJ&F: 65.5 -> 66.9Improved
use both first and previous frame as references (ablation)UniRef++-R50 · Youtube-VOS2018G: 80.0 -> 81.9Improved
add mask reference for RVOS on Ref-DAVIS17 (UniRef++ vs language-only UniRef)UniRef++-L · Ref-DAVIS17J&F: UniRef-R50 63.5 vs UniRef++-R50 62.5Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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