Research map / Video / Video segmentation and tracking
Online Mutual Foreground Segmentation for Multispectral Stereo Videos
Pierre-Luc St-Charles, Guillaume-Alexandre Bilodeau, Robert Bergevin
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 mutual stereo registration and segmentation | unsupervised video segmentation baseline · VAP 2016 segmentation | F1: 0.760 -> 0.866 | Improved |
| add the shape term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | mean disparity error (px): 8.71 -> 3.21 | Improved |
| add the contour term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.624 -> 0.866 | Improved |
| add the color term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.822 -> 0.866 | Improved |
| add the appearance term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.851 -> 0.866 | Improved |
| add saliency maps (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.856 -> 0.866 | Improved |
| add the uniqueness term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.865 -> 0.866 | No clear effect |
| add the temporal term (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.855 -> 0.866 | Improved |
| add the initial LWIR segmentation mask (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | mean disparity error (px): 10.82 -> 3.21 | Improved |
| add the initial visible segmentation mask (ablation) | proposed method · VAP 2016 + Bilodeau 2014 (overall) | F1: 0.800 -> 0.866 | Improved |
| allow deferred output with 3 layers instead of real-time (ablation) | proposed method · VAP 2016 segmentation | F1: 0.866 -> 0.870 | No clear effect |
| increase temporal pipeline from 2 to 5 layers (real-time) (ablation) | proposed method · VAP 2016 segmentation | F1: 0.863 -> 0.868 | No clear effect |
Closest papers by meaning
- A Multilayer-Based Framework for Online Background Subtraction With Freely Moving Cameras (ICCV 2017 · similarity 0.86)
- Stereo Any Video: Temporally Consistent Stereo Matching (ICCV 2025 · similarity 0.85)
- Match-Stereo-Videos: Bidirectional Alignment for Consistent Dynamic Stereo Matching (ECCV 2024 · similarity 0.85)
- Temporally Consistent Stereo Matching (ECCV 2024 · similarity 0.85)
- FOCUS: Towards Universal Foreground Segmentation (AAAI 2025 · similarity 0.84)
- Unleashing Multispectral Video's Potential in Semantic Segmentation: A Semi-supervised Viewpoint and New UAV-View Benchmark (NeurIPS 2024 · similarity 0.83)
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.
