Research map / Video / Video segmentation and tracking
Primary Object Segmentation in Videos via Alternate Convex Optimization of Foreground and Background Distributions
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use ACO instead of [37] | unsupervised video object segmentation · SegTrack, SegTrack v2, VidSeg (average) | IoU: 56.02 -> 70.24 | Helped |
| use ACO instead of [29] | unsupervised video object segmentation · SegTrack, SegTrack v2, VidSeg (average) | IoU: 51.24 -> 70.24 | Helped |
| add the forward pass to initial per-frame estimates (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 44.31 -> 64.80 | Helped |
| add the backward pass after the forward pass (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 64.80 -> 67.71 | Helped |
| add pixel-wise refinement after the backward pass (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 67.71 -> 70.24 | Helped |
| add Markov energy to spatiotemporal and antagonistic energies (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 52.92 -> 67.71 | Helped |
| add spatiotemporal energy to Markov and antagonistic energies (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 11.30 -> 67.71 | Helped |
| add antagonistic energy to Markov and spatiotemporal energies (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 57.87 -> 67.71 | Helped |
| use only spatiotemporal energy instead of all three terms (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 67.71 -> 52.81 | Hurt |
| use spatiotemporal and antagonistic energies without Markov energy (ablation) | ACO · SegTrack, SegTrack v2, VidSeg (average) | IoU: 67.71 -> 52.92 | Hurt |
Closest papers by meaning
- Primary Object Segmentation in Videos Based on Region Augmentation and Reduction (CVPR 2017 · similarity 0.87)
- Reciprocal Transformations for Unsupervised Video Object Segmentation (CVPR 2021 · similarity 0.87)
- Iteratively Selecting an Easy Reference Frame Makes Unsupervised Video Object Segmentation Easier (AAAI 2022 · similarity 0.85)
- Learning Unsupervised Video Object Segmentation Through Visual Attention (CVPR 2019 · similarity 0.85)
- M^3-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object Segmentation (CVPR 2025 · similarity 0.85)
- Online Video Object Segmentation via Convolutional Trident Network (CVPR 2017 · 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.