Research map / Video / Video segmentation and tracking
Temporal Superpixels Based on Proximity-Weighted Patch Matching
Se-Ho Lee, Won-Dong Jang, Chang-Su Kim
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 |
|---|---|---|---|
| replace SLIC with the proposed superpixel algorithm in a semi-supervised VOS method | VOS method of [10] · SegTrack | average IoU: 0.532 -> 0.579 | Improved |
| post-process ACO results with temporal superpixels | ACO · SegTrack | average IoU: 0.574 -> 0.595 | Improved |
| post-process FOS results with temporal superpixels | FOS · SegTrack | average IoU: 0.438 -> 0.446 | Improved |
| add PPM motion estimation instead of DeepFlow-averaged flow (ablation) | TS-PPM · SegTrack (400 supervoxels) | UE2D: 6.3% better | Improved |
| add PPM motion estimation instead of DeepFlow-averaged flow (ablation) | TS-PPM · SegTrack (400 supervoxels) | UE3D: 8.6% better | Improved |
| add PPM motion estimation instead of DeepFlow-averaged flow (ablation) | TS-PPM · SegTrack (400 supervoxels) | SA3D: 4.6% better | Improved |
| add PPM motion estimation instead of DeepFlow-averaged flow (ablation) | TS-PPM · SegTrack (other metrics) | other metrics average: about 0.5% on average | Improved |
| add optional backward refinement (ablation) | TS-PPM · SegTrack | SA2D, SA3D, BR2D, BR3D, EV | Improved |
| add optional backward refinement (ablation) | TS-PPM · SegTrack | UE2D, UE3D | No clear effect |
| add optional backward refinement (ablation) | TS-PPM · SegTrack | mean duration | No clear effect |
| add optional backward refinement (ablation) | TS-PPM · Run-time per frame | time (s): 0.29 -> 0.38 | Hurt |
| use proposed superpixels instead of CCS | superpixel algorithm · BSDS500 (200 segments) | UE: 6.3% lower UE | Improved |
Closest papers by meaning
- Contour-Constrained Superpixels for Image and Video Processing (CVPR 2017 · similarity 0.87)
- Occlusion-Aware Method for Temporally Consistent Superpixels (TPAMI 2018 · similarity 0.87)
- Extract and Merge: Superpixel Segmentation with Regional Attributes (ECCV 2020 · similarity 0.86)
- Generalized Closed-Form Formulae for Feature-Based Subpixel Alignment in Patch-Based Matching (IJCV 2025 · similarity 0.85)
- Spatiotemporal GMM for Background Subtraction with Superpixel Hierarchy (TPAMI 2017 · similarity 0.85)
- Learning Invariant Inter-pixel Correlations for Superpixel Generation (AAAI 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.
