Research map / Video / Video segmentation and tracking
Video Object Segmentation without Temporal Information
Kevis-Kokitsi Maninis, Sergi Caelles, Yuhua Chen, Jordi Pont-Tuset, Laura Leal-Taixé, Daniel Cremers, Luc Van Gool
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 |
|---|---|---|---|
| fine-tune on first-frame mask (OneShot) (ablation) | ImageNet network · DAVIS 2016 val | J&F mean: 18.9 -> 65.6 | Improved |
| add parent network pre-training (ablation) | OneShot · DAVIS 2016 val | J&F mean: 65.6 -> 77.8 | Improved |
| add semantics and conditional classifier (ablation) | +Parent · DAVIS 2016 val | J&F mean: 77.8 -> 86.1 | Improved |
| add superpixel snapping (ablation) | +Semantics · DAVIS 2016 val | J&F mean: 86.1 -> 85.4 | No clear effect |
| add contour recovery (ablation) | +Superpixels · DAVIS 2016 val | J&F mean: 85.4 -> 86.5 | Improved |
| add semantics (ablation) | +Parent · DAVIS 2016 val | T temporal instability: 33.5 -> 20.2 | Improved |
| use Mask-RCNN instead of MNC for instance segmentation (ablation) | OSVOS-S · DAVIS 2016 val | J&F mean: 83.5 -> 86.5 | Improved |
| use Mask-RCNN instead of FCIS for instance segmentation (ablation) | OSVOS-S · DAVIS 2017 test-dev | J&F: 53.7 -> 57.5 | Improved |
| add semantic instance prior to OSVOS | OSVOS · DAVIS 2016 val | J&F mean: 80.2 -> 86.5 | Improved |
| train parent network with 200 frames instead of full 2079 (ablation) | OSVOS-S parent · DAVIS 2016 val | J mean: 84.9 -> 85.6 | No clear effect |
| train parent network with only 100 frames (ablation) | OSVOS-S parent · DAVIS 2016 val | J mean: 85.6 -> 82.3 | Hurt |
| automatic instance selection instead of oracle selection | OSVOS-S · DAVIS 2016 val | J&F mean: 82.6 -> 82.4 | No clear effect |
Closest papers by meaning
- One-Shot Video Object Segmentation (CVPR 2017 · similarity 0.96)
- Make One-Shot Video Object Segmentation Efficient Again (NeurIPS 2020 · similarity 0.91)
- XMem++: Production-level Video Segmentation From Few Annotated Frames (ICCV 2023 · similarity 0.90)
- Video Object Segmentation Using Space-Time Memory Networks (ICCV 2019 · similarity 0.90)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 2025 · similarity 0.90)
- A Survey on Deep Learning Technique for Video Segmentation (TPAMI 2022 · similarity 0.90)
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.
