Research map / Video / Video segmentation and tracking
Robust Online Matrix Factorization for Dynamic Background Subtraction
Hongwei Yong, Deyu Meng, Wangmeng Zuo, Lei 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 temporal smoothness (TV) on foreground to OMoGMF | OMoGMF · Li dataset average | F-measure (%): 70.54 -> 77.19 | Improved |
| add MoG noise modeling per frame (OMoGMF) vs fixed-noise online methods | online subspace methods · Li dataset average | F-measure (%): 70.54 vs 64.61 OPRMF and 63.05 GRASTA | Improved |
| add transformation operator (t-OMoGMF) (ablation) | OMoGMF · Li dataset with artificial transformations | F-measure (%): 40.07 -> 73.27 | Improved |
| add transformation operator (t-OMoGMF vs RASL) | RASL · Real camera-jitter videos | F-measure (%): 65.35 -> 70.60 | Improved |
| add transformation operator | OMoGMF · Li synthetic jitter | speed: more than 5 times faster than RASL | Improved |
| lower pixel sub-sampling rate to 1% for OMoGMF (ablation) | OMoGMF · Sub-sampled Li videos | F-measure (%): 70.34 -> 69.82 | No clear effect |
| lower pixel sub-sampling rate to 1% for OMoGMF (ablation) | OMoGMF · Sub-sampled Li videos | FPS: 57.6 -> 273.4 | Improved |
| lower pixel sub-sampling rate to 1% for GRASTA (ablation) | GRASTA · Sub-sampled Li videos | F-measure (%): 63.07 -> 57.16 | Hurt |
| use OMoGMF instead of GRASTA at 1% sub-sampling | GRASTA · Sub-sampled Li videos | F-measure (%): 57.16 -> 69.82 | Improved |
| add TV smoothness to OMoGMF (ablation) | OMoGMF · escalator FPS | FPS: 99.6 -> 18.5 | Hurt |
| OMoGMF vs incPCP speed | incPCP · escalator FPS | FPS: 274.5 vs 99.6 | Hurt |
Closest papers by meaning
- Spatial-Temporal Gaussian Scale Mixture Modeling for Foreground Estimation (AAAI 2020 · similarity 0.88)
- 4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization (NeurIPS 2024 · similarity 0.84)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.82)
- Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection (NeurIPS 2025 · similarity 0.82)
- EigenGS Representation: From Eigenspace to Gaussian Image Space (CVPR 2025 · similarity 0.82)
- Efficient Gaussian Splatting for Monocular Dynamic Scene Rendering via Sparse Time-Variant Attribute Modeling (AAAI 2025 · similarity 0.82)
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.
