Research map / Detection and segmentation / Tracking
Non-Rigid Object Tracking via Deformable Patches Using Shape-Preserved KCF and Level Sets
Xin Sun, Ngai-Man Cheung, Hongxun Yao, Yiluan Guo
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 |
|---|---|---|---|
| use deformable level-set patches instead of fixed rectangular patches (RPT) | patch-based KCF tracker · VOT2014 (25 sequences) | correctly tracked frames (%): 65.4868 -> 71.0828 | Improved |
| use deformable level-set patches instead of fixed rectangular patches (RPT) | patch-based KCF tracker · VOT2016 (60 sequences) | correctly tracked frames (%): 54.0174 -> 56.7567 | Improved |
| use weighted local patches instead of a single global boosting model with level sets (SLSM) | contour tracker · VOT2014 (25 sequences) | correctly tracked frames (%): 56.4132 -> 71.0828 | Improved |
| use weighted local patches instead of a single global boosting model with level sets (SLSM) | contour tracker · VOT2016 (60 sequences) | correctly tracked frames (%): 47.8365 -> 56.7567 | Improved |
| use local patch models instead of HoughTrack global model | contour tracker · VOT2014 (25 sequences) | correctly tracked frames (%): 51.7184 -> 71.0828 | Improved |
| use deformable patches instead of distribution field tracker | bounding box tracker · VOT2016 (60 sequences) | correctly tracked frames (%): 42.0626 -> 56.7567 | Improved |
| use deformable patches versus PixelTrack | bounding box tracker · VOT2016 (60 sequences) | correctly tracked frames (%): 40.3302 -> 56.7567 | Improved |
| proposed tracker versus all compared trackers per sequence | proposed tracker · VOT2014 sequences | correctly tracked frames (%): 13 out of 25 | Mixed |
| proposed versus RPT on david | proposed tracker · VOT2014 david | correctly tracked frames (%): 100 -> 78.57 | Hurt |
| proposed versus RPT on trellis | proposed tracker · VOT2014 trellis | correctly tracked frames (%): 100 -> 39.72 | Hurt |
| proposed versus RPT on diving | proposed tracker · VOT2014 diving | correctly tracked frames (%): 15.98 -> 100 | Improved |
| proposed versus RPT on jogging | proposed tracker · VOT2014 jogging | correctly tracked frames (%): 22.48 -> 100 | Improved |
Closest papers by meaning
- Structural Correlation Filter for Robust Visual Tracking (CVPR 2016 · similarity 0.86)
- Robust Visual Tracking via Local-Global Correlation Filter (AAAI 2017 · similarity 0.85)
- Learning Patch-Based Dynamic Graph for Visual Tracking (AAAI 2017 · similarity 0.85)
- Textureless Deformable Object Tracking With Invisible Markers (TPAMI 2024 · similarity 0.85)
- Tracking by 3D Model Estimation of Unknown Objects in Videos (ICCV 2023 · similarity 0.85)
- Robust Multi-Body Feature Tracker: A Segmentation-Free Approach (CVPR 2016 · 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.
