Mean Shift: A Robust Approach Toward Feature Space Analysis

By Dorin Comaniciu, Peter Meer. 2002

12,922 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Statistics and Probability, Algorithms & Complexity

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Statistics and Probability

In Algorithms & Complexity

Citations in the corpus

3D Instance Segmentation of MVS Buildings (2021, methodology): “Many clustering methods are available in the literature, such as Mean-shift [55] and K-means [56], but none of them is suitable for our multi-view scenario.”
Fast Edge Detection Using Structured Forests (2014, neutral): “81 Mean Shift [9] .”
Towards Better Analysis of Deep Convolutional Neural Networks (2016, methodology): “In CNNVis, we employ two widely used clustering methods, K-Means [36] (parametric clustering) and MeanShift [8] (nonparametric clustering).”
IM2GPS: estimating geographic information from a single image (2008, methodology): “One way to operationalize this is to consider the major modes of the distribution by performing mean-shift [3] clustering on the geolocations of the matches.”
Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics (2017, background): “Secondly, it does not assume a canonical instance size or density like discretised binning approaches [12].”
DeepEdge: A multi-scale bifurcated deep network for top-down contour detection (2014, neutral): “ODS OIS AP Canny [7] 0.6 0.63 0.58 Mean Shift [8] 0.64 0.68 0.56 Ncuts [26] 0.64 0.68 0.45 Felz., Hutt.”
MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance (2019, background): “) (Wand and Jones, 1994) can be illustrated as: One typical solution for KDE clustering to minimize f ( z ) is taking Mean Shift (Comaniciu and Meer, 2002), deﬁned as: γ ij k (cid:48) ( d ( v j , z i,j )) ∂d ( v j , z i,j ) ∂ v Firstly, v τ +1 j can be updated while c τ +1 ij is ﬁxed: )) z i,j i,j k”
Recurrent Pixel Embedding for Instance Grouping (2017, methodology): “Mean-shift and closely related algorithms [26, 13, 14, 15] use kernel density estimation to approximate the probability density from a set of samples and then perform clustering on the input data by assigning or moving each sample to the nearest mode (local maxima).”
Multi-Scale Representation Learning on Proteins (2022, methodology): “Popular graph-based segmentation tools such as Felzenszwalb and Huttenlocher [2004, FH], mean shift [Comaniciu and Meer, 2002], and watershed [Vincent and Soille, 1991], however, produce non-compact superpixels of irregular sizes and shapes.”
ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data (2019, methodology): “Popular image segmentation algorithm in remote sensing include watershed segmentation [55], multi-resolution segmentation [4] and mean-shift segmentation [12].”
A Novel Performance Evaluation Methodology for Single-Target Trackers (2015, background): “(which is a Mean Shift tracker [70] with an isotropic kernel”
Nuclei Segmentation Using Marker-Controlled Watershed, Tracking Using Mean-Shift, and Kalman Filter in Time-Lapse Microscopy (2006, background): “Digital Object Identifier 10.1109/TCSI.2006.884469 interest in realizing the full potential of time-lapse microscopy in biological research or drug discovery [25], [26].”
Improved fast gauss transform and efficient kernel density estimation (2003, methodology): “We do not perform any postprocessing procedure as in [5].”
Human Instance Segmentation from Video using Detector-based Conditional Random Fields (2011, methodology): “Multiple segmentations are generated for each frame using the meanshift algorithm [3], varying (spatial and range) parameters, hs,hr.”