TurboPixels: Fast Superpixels Using Geometric Flows

By Alex Levinshtein, Adrian Stere, Kiriakos N. Kutulakos, David J. Fleet, Sven J. Dickinson, Kaleem Siddiqi. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009

1,235 citations recorded by the source corpus.

Subfields: Biophysics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/a3269505b943d3549a82c4eef23d9e29cea3be11

How researchers see this paper

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

In Biophysics

In Computer Vision and Pattern Recognition

Citations in the corpus

Robust Superpixel Tracking (2014, methodology): “In particular, superpixels have been applied to image segmentation and object recognition with demonstrated success [20], [23], [24], [27], [28].”
Superpixel tracking (2011, background): “In particular, superpixels have been one of the most promising representations with demonstrated success in image segmentation and object recognition [18] [15] [22] [12] [17].”
Image-based plant phenotyping with incremental learning and active contours (2014, methodology): “, 2011) method which relies on the temperature maximization of the anisotropic heat diffusion formulation on a graph representation of the image (an initial super-pixel segmentation is obtained using (Levinshtein et al., 2009)); and finally, although it is not a true segmentation method, we also use”
Fast Object Segmentation in Unconstrained Video (2013, methodology): “Currently, we use TurboPixels as superpixels [15] (1.”
Towards automatic polyp detection with a polyp appearance model (2012, neutral): “We compare our final segmentation results with related techniques such as normalized cuts [27] and turbo pixels [28].”
Beyond pixels: A comprehensive survey from bottom-up to semantic image segmentation and cosegmentation (2015, neutral): “TurboPixel [63] deforms the initial spatial grid to compact and regular regions by using geometric flow”
Robust Nucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review (2016, methodology): “[309] have used a structured SVM classifier to pick the optimal superpixels, which are created by Turbopixels [310].”
SLIC Superpixels Compared to State-of-the-Art Superpixel Methods (2012, methodology): “32 ) [ 15], where N is the number of pixels.”
Superpixels and Polygons Using Simple Non-iterative Clustering (2017, methodology): “We compare SNIC1 to SLIC [7] as well as several stateof-the-art superpixel methods: NCUTS [28], MST [13], TURBO [16], SEEDS [30], ERS [18], and LSC [17].”
Training Object Class Detectors from Eye Tracking Data (2014, methodology): “We start by over-segmenting each image I into superpixels S using the Turbopixels method [29].”
Locally Orderless Tracking (2012, methodology): “Specifically, target and candidate patches are represented by signatures which are generated from superpixels computed using TurboPixels [26] clustering built in a region-of-intrest (ROI) which supports all the particles.”
Distributed cosegmentation via submodular optimization on anisotropic diffusion (2011, methodology): “Bellow, we begin with the segmentation of a single image to illustrate the basic behavior of the algorithm.”
Ontology-Based Semantic Image Segmentation Using Mixture Models and Multiple CRFs (2016, methodology): “Several unsupervised segmentation algorithms have hence been used for higher level region-based feature representation such as JSEG [4], mean shift [5], graph-based segmentation [6], [7], quick shift [8], TurboPixel [9] and SLIC superpixel [10].”
Lazy Random Walks for Superpixel Segmentation (2014, background): “The second category of superpixel algorithms considers the compactness constrains, such as normalized cuts [7], lattice cut [16], TurboPixels [17], and graph cut [19] approaches.”
Image Segmentation by Cascaded Region Agglomeration (2013, methodology): “Superpixels In superpixel regime, we compare ISCRA to SLIC, ERS, OWT-UCM, and Hoiem et al. SLIC and ERS have been shown in [1, 14] to outperform other methods, including MeanShift [4], Turbopixels [13], graph-based methods [18, 6, 22], and others [17, 8], on at least some of the data sets, and we do”