Research map / Video / Video segmentation and tracking
LVOS: A Benchmark for Long-term Video Object Segmentation
Lingyi Hong, Wenchao Chen, Zhongying Liu, Wei Zhang, Pinxue Guo, Zhaoyu Chen, Wenqiang Zhang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add local memory to reference memory (ablation) | DDMemory · LVOS validation | J&F: 44.2 -> 57.9 | Improved |
| add global memory to reference and local memory (ablation) | DDMemory · LVOS validation | J&F: 57.9 -> 61.9 | Improved |
| add reference memory to global and local memory (ablation) | DDMemory · LVOS validation | J&F: 54.9 -> 61.9 | Improved |
| use local memory alone (ablation) | DDMemory · LVOS validation | J&F: 61.9 -> 18.3 | Hurt |
| finetune on LVOS training set | existing VOS models and DDMemory · LVOS validation | J&F: 60.7 -> 61.9 (DDMemory); 59.4 -> 60.9 (AOT-L) | Improved |
| provide oracle bounding box | DDMemory · LVOS validation | J&F: 61.9 -> 70.2 | Improved |
| provide oracle mask for memory update | DDMemory · LVOS validation | J&F: 61.9 -> 82.7 | Improved |
| evaluate DDMemory on short videos with global memory | DDMemory · DAVIS 2017 validation | J&F: 84.2 vs XMem 86.2 | No clear effect |
| use compressed memory bank or online adaptation models on scale variation | VOS models on LVOS attributes · LVOS validation attributes | J: LRA 44.3 (AOT-B) vs FM 54.3 | Hurt |
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.
- XMem — XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model (ECCV 2022) · better on 3 of 5 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
- LVOS: A Benchmark for Large-Scale Long-Term Video Object Segmentation (TPAMI 2025 · similarity 0.98)
- OneVOS: Unifying Video Object Segmentation with All-in-One Transformer Framework (ECCV 2024 · similarity 0.91)
- Long-RVOS: A Comprehensive Benchmark for Long-term Referring Video Object Segmentation (CVPR 2026 · similarity 0.90)
- FEELVOS: Fast End-To-End Embedding Learning for Video Object Segmentation (CVPR 2019 · similarity 0.90)
- Point-VOS: Pointing Up Video Object Segmentation (CVPR 2024 · similarity 0.90)
- RMem: Restricted Memory Banks Improve Video Object Segmentation (CVPR 2024 · similarity 0.88)
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.
