Research map / Video / Video segmentation and tracking
LIBSVX: A Supervoxel Library and Benchmark for Early Video Processing
Chenliang Xu, Jason J. Corso
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 TSP supervoxels instead of other methods | supervoxel benchmark · UE3D (all datasets) | 3D undersegmentation error | Improved |
| use GBH supervoxels instead of other methods | supervoxel benchmark · BRD (all datasets) | boundary recall distance | Improved |
| use SWA supervoxels instead of other methods | supervoxel benchmark · all datasets | explained variation | Improved |
| use TSP (optical-flow based) supervoxels | supervoxel benchmark · Middlebury Flow and flow dataset | label consistency | Improved |
| use GBH supervoxels instead of other methods | supervoxel benchmark · all datasets | temporal extent | Improved |
| use TSP supervoxels instead of other methods | supervoxel benchmark · all datasets | mean size variation | Improved |
| use streaming GBH instead of full-video GBH | supervoxel classification · CamVid | classification accuracy | No clear effect |
| use faster methods (GB, Mean Shift) instead of GBH/SWA/TSP | supervoxel benchmark · all datasets | quality on human-annotation metrics | Hurt |
| use supervoxels with appearance-only features instead of pixel-level random forests with geometric cues | CamVid labeling · CamVid | global vs per-class accuracy: better global pixel accuracy but worse average per-class accuracy than 53.0 / 69.1 | Mixed |
Closest papers by meaning
- Fast Computation of Content-Sensitive Superpixels and Supervoxels Using Q-Distances (ICCV 2019 · similarity 0.89)
- Superpixel Segmentation With Fully Convolutional Networks (CVPR 2020 · similarity 0.89)
- Feature-Aware Uniform Tessellations on Video Manifold for Content-Sensitive Supervoxels (TPAMI 2020 · similarity 0.88)
- Content-Sensitive Supervoxels via Uniform Tessellations on Video Manifolds (CVPR 2018 · similarity 0.88)
- Spatiotemporal GMM for Background Subtraction with Superpixel Hierarchy (TPAMI 2017 · similarity 0.87)
- Hierarchical Histogram Threshold Segmentation - Auto-terminating High-detail Oversegmentation (CVPR 2024 · 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.
