Research map / Learning theory and algorithms / Clustering
On the Consistency of Quick Shift
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Quick Shift with τ < r_M/2 and the stated bandwidth conditions | Quick Shift · Density mode recovery | Hausdorff distance: d(\mathcal{M}, \widehat{\mathcal{M}}) = \tilde{O}(n^{-1/(4+d)}) | Helped |
| Tune τ to merge modes that are not maximal in their τ-radius neighborhood | Quick Shift · Density mode recovery with segmentation | Mode recovery guarantee: ||x_0 - \hat{x} ||^2 < C\left((\log n)^{4/\rho} h^{2} + \sqrt{\frac{\log n}{n \cdot h^d}}\right) | Mixed |
| Use Quick Shift edges with τ < r_s/2 | Quick Shift · Density cluster assignments | Separation of assigned clusters | Helped |
| Build level-set components from Quick Shift and link nearby components | Quick Shift Cluster Tree Estimator · Density cluster tree | Merge distortion consistency | Helped |
| Apply Quick Shift to estimate conditional-density modes | Quick Shift Modal Regression · Conditional density modes | Hausdorff distance: d_H(\mathcal{M}_x, \widehat{\mathcal{M}}_x) \rightarrow 0 \text{ as } n\rightarrow \infty | Helped |
Closest papers by meaning
- TANGO: Clustering with Typicality-Aware Nonlocal Mode-Seeking and Graph-Cut Optimization (ICML 2025 · similarity 0.82)
- ReTaSA: A Nonparametric Functional Estimation Approach for Addressing Continuous Target Shift (ICLR 2024 · similarity 0.81)
- Graph–Smoothed Bayesian Black-Box Shift Estimator and Its Information Geometry (NeurIPS 2025 · similarity 0.81)
- To discretize continually: Mean shift interacting particle systems for Bayesian inference (NeurIPS 2026 · similarity 0.81)
- Contrastive Mean-Shift Learning for Generalized Category Discovery (CVPR 2024 · similarity 0.81)
- Shifting the Paradigm: A Diffeomorphism Between Time Series Data Manifolds for Achieving Shift-Invariancy in Deep Learning (ICLR 2025 · similarity 0.81)
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.