Research map / Video / Video segmentation and tracking
No Labels, No Look-Ahead: Unsupervised Online Video Stabilization with Classical Priors
Tao Liu, Kan Ren, Gang Wan, Shibo Wen
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 motion propagation module (ablation) | unsupervised online stabilizer · five datasets | D and PSNR | Improved |
| add online trajectory smoothing (ablation) | unsupervised online stabilizer · five datasets | structural stability (D) | Improved |
| add both motion propagation and smoothing (ablation) | unsupervised online stabilizer · five datasets | all metrics | Improved |
| add keypoint consistency loss (ablation) | unsupervised online stabilizer · five datasets | D, PSNR | Improved |
| add temporal regularization loss (ablation) | unsupervised online stabilizer · five datasets | D (stability) | Improved |
| use multi-homography prior instead of single homography (ablation) | unsupervised online stabilizer · five datasets | D, SSIM | Improved |
| add multi-detector keypoint collaboration instead of raw keypoints (ablation) | unsupervised online stabilizer · five datasets | D | Improved |
| shorten smoothing window from 7 to 5 frames (ablation) | OnlineSmoother · five datasets | stability vs fidelity | Mixed |
| lengthen smoothing window from 7 to 9 frames (ablation) | OnlineSmoother · five datasets | all metrics | No clear effect |
| use the proposed causal method versus NNDVS | unsupervised online stabilizer · UAV-Test | stability score: 0.84 -> 0.89 | Improved |
| use the proposed method versus MeshFlow | unsupervised online stabilizer · DeepStab | stability score: 0.86 -> 0.85 | Hurt |
Closest papers by meaning
- Minimum Latency Deep Online Video Stabilization (ICCV 2023 · similarity 0.87)
- Minimum Latency Deep Online Video Stabilization and Its Extensions (TPAMI 2024 · similarity 0.86)
- Harnessing Meta-Learning for Improving Full-Frame Video Stabilization (CVPR 2024 · similarity 0.85)
- Harnessing Meta-Learning for Controllable Full-Frame Video Stabilization (TPAMI 2026 · similarity 0.85)
- Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks (NeurIPS 2025 · similarity 0.83)
- $h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement (NeurIPS 2026 · 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.
