Research map / Video / Video segmentation and tracking
Robust Promptable Video Object Segmentation
Sohyun Lee, Yeho Gwon, Lukas Hoyer, Konrad Schindler, Christos Sakaridis, Suha Kwak
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 memory conditioning of the gate (ablation) | MoGA · MVSeg-adv | J&F: 69.6 -> 70.9 | Improved |
| add per-object conditioning on top of memory conditioning (ablation) | MoGA · MVSeg-adv | J&F: 70.9 -> 71.8 | Improved |
| use MoGA instead of LoRA | SAM2 · MVSeg-adv | J&F: 70.9 -> 71.8 | Improved |
| use MoGA with 1.1M params instead of full fine-tuning of 80.9M params | SAM2 · MVSeg-adv | J&F: 71.5 -> 71.8 | Improved |
| add MoGA to SAM2 | SAM2 · ACDC-Video | J&F: 63.5 -> 64.5 | Improved |
| add MoGA to SAM2 | SAM2 · YouTube-VOS-C | J&F: 78.7 -> 79.9 | Improved |
| add MoGA to SAM2 | SAM2 · YouTube-VOS (clean) | J&F: 82.2 -> 82.6 | No clear effect |
| add MoGA to SAM2 on long videos | SAM2 · MVSeg-adv long videos (~42s) | J&F: 52.3 -> 56.2 | Improved |
| apply GaRA frame-wise before SAM2 | SAM2 · YouTube-VOS-C | J&F: 78.7 -> 78.1 | Hurt |
| apply restoration (AirNet) before SAM2 | SAM2 · ACDC-Video | J&F: 63.5 -> 59.8 | Hurt |
| increase adapter rank from 128 to 256 or 512 (ablation) | MoGA · YouTube-VOS-C | J&F: 79.9 -> 79.8, 79.7 | Hurt |
| reduce adapter rank from 128 to 32 (ablation) | MoGA · YouTube-VOS-C | J&F: 79.9 -> 79.3 | 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.
- SAM2 — SAM 2: Segment Anything in Images and Videos (ICLR 2025) · better on 7 of 7 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
- Learning Fast and Robust Target Models for Video Object Segmentation (CVPR 2020 · similarity 0.87)
- Long-RVOS: A Comprehensive Benchmark for Long-term Referring Video Object Segmentation (CVPR 2026 · similarity 0.87)
- LVOS: A Benchmark for Large-Scale Long-Term Video Object Segmentation (TPAMI 2025 · similarity 0.87)
- RMem: Restricted Memory Banks Improve Video Object Segmentation (CVPR 2024 · similarity 0.86)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 2025 · similarity 0.85)
- LiVOS: Light Video Object Segmentation with Gated Linear Matching (CVPR 2025 · similarity 0.85)
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.
