Research map / Video / Video segmentation and tracking
Fast Computation of Content-Sensitive Superpixels and Supervoxels Using Q-Distances
Zipeng Ye, Ran Yi, Minjing Yu, Yong-Jin Liu, Ying He
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 Dijkstra's priority queue with a FIFO queue (q-distance) (ablation) | GCVT superpixels (IMSLIC) · BSDS500 (481 x 321 images) | running time speedup: 6-8 times faster | Improved |
| use q-distance with centroid-free Lloyd refinement and K-means++ initialization instead of Dijkstra GCVT of IMSLIC | IMSLIC · BSDS500 | under-segmentation error | Improved |
| select more cluster centers K (ablation) | qd-CSS q-distances · image example (K = 20, 100, 300) | wrongly-labelled pixels (q-distance differs from shortest distance) | Improved |
| compare with deep SSN and SEAL on UE and BR | qd-CSS · BSDS500 | under-segmentation error and boundary recall | Hurt |
| compare with SEEDS on boundary recall | qd-CSS · BSDS500 | boundary recall | Hurt |
| compare with ETPS and SLIC on running time | qd-CSS · BSDS500 | running time: two times slower | Hurt |
| compare with Yi-CSS and 6 other supervoxel methods | qd-CSS · BuffaloXiph | UE3D, BRD, compactness | Improved |
| use qd-CSS supervoxels in foreground propagation | qd-CSS · Youtube-Objects (10 classes) | average F measure | Improved |
Closest papers by meaning
- Manifold SLIC: A Fast Method to Compute Content-Sensitive Superpixels (CVPR 2016 · similarity 0.92)
- Intrinsic Manifold SLIC: A Simple and Efficient Method for Computing Content-Sensitive Superpixels (TPAMI 2017 · similarity 0.91)
- Content-Sensitive Supervoxels via Uniform Tessellations on Video Manifolds (CVPR 2018 · similarity 0.90)
- Feature-Aware Uniform Tessellations on Video Manifold for Content-Sensitive Supervoxels (TPAMI 2020 · similarity 0.89)
- LIBSVX: A Supervoxel Library and Benchmark for Early Video Processing (IJCV 2016 · similarity 0.89)
- Learning Invariant Inter-pixel Correlations for Superpixel Generation (AAAI 2024 · similarity 0.84)
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.
