Vectorized image segmentation via trixel agglomeration

Prasad, Lakshman; Skourikhine, Alexei N.

doi:10.1016/j.patcog.2005.10.014

Cited by 37 publications

(15 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second segmentation algorithm is the dual contour/region segmentation of Prasad and Skourikine [23]( figure 3(b)). The approach consists in an edge detection step followed by a Constrained Delaunay Triangulation (CDT) to close the region contours.…”

Section: Image Segmentationmentioning

confidence: 99%

Segmentation-based multi-class semantic object detection

Vieux

Benois-Pineau

Domenger

et al. 2010

Multimed Tools Appl

View full text Add to dashboard Cite

In this paper we study the problem of the detection of semantic objects from known categories in images. Unlike existing techniques which operate at the pixel or at a patch level for recognition, we propose to rely on the categorization of image segments. Recent work has highlighted that image segments provide a sound support for visual object class recognition. In this work, we use image segments as primitives to extract robust features and train detection models for a predefined set of categories. Several segmentation algorithms are benchmarked and their performances for segment recognition are compared. We then propose two methods for enhancing the segments classification, one based on the fusion of the classification results obtained with the different segmentations, the other one based on the optimization of the global labelling by correcting local ambiguities between neighbor segments. We use as a benchmark the Microsoft MSRC-21 image database and show that our method competes with the current state-of-the-art.

show abstract

Section: Image Segmentationmentioning

confidence: 99%

Segmentation-based multi-class semantic object detection

Vieux

Benois-Pineau

Domenger

et al. 2010

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…We will use our vectorized image segmentation framework [5,6,7] to obtain our initial edge-based seed segmentation.…”

Section: Vectorized Image Segmentationmentioning

confidence: 99%

“…In contrast, we did not vary any parameters of our method across the images tested on in this paper. The Canny edge detection parameters that govern the initial segmentation as described in [7] were fixed at sigma = 1 and the high and low hysteresis thresholds were fixed at 0.6 and 0.0, respectively. Both the multiscale Ncuts and the mean shift methods yield over-segmentations of salient objects of interest compared to our method.…”

Section: Comparisonmentioning

confidence: 99%

“…In this paper we will present an alternate approach to hierarchical image segmentation based on an earlier work of ours [5,6,7] to obtain an initial fine scale 'seed' segmentation that faithfully conforms to image contours. We then iteratively group these polygons based on their size distributions and spectral proximities to yield a hierarchy of image segmentations wherein each polygon at each level is an aggregate of one or more polygons at the previous finer level.…”

Section: Introductionmentioning

confidence: 99%

“…Section 4 discusses how object and texture polygons are distinguishable from their contour characteristics. Table 1: Boolean perceptual edge filters [7].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hierarchical image segmentation by polygon grouping

Prasad

Swaminarayan

2008

2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

Self Cite

View full text Add to dashboard Cite

We present a simple cascading algorithm for rapid hierarchical image segmentation based on perceptually driven contour completion and scene statistics. We start with an initial fine-scale segmentation of an image obtained by perceptual completion of partial contours into polygonal regions using region-contour correspondences established by Delaunay triangulation of edge pixels. The resulting polygon size distribution is analyzed for a dominant mode of granularity of the image. Polygons whose sizes are less than or equal to this granularity are merged with their spectrally closest neighbors to obtain the next level of polygonal segments in the hierarchy. The iterative application of this process precipitates textured regions as polygons with highly convolved boundaries and helps distinguish them from objects which typically have more regular boundaries.

show abstract

A Color Channel Based Analysis on Image Tessellation

Kibar,

Genç

2023

Engineering Cyber-Physical Systems and Critical Infrastructures

View full text Add to dashboard Cite

Vectorized image segmentation via trixel agglomeration

Cited by 37 publications

References 7 publications

Segmentation-based multi-class semantic object detection

Segmentation-based multi-class semantic object detection

Hierarchical image segmentation by polygon grouping

A Color Channel Based Analysis on Image Tessellation

Contact Info

Product

Resources

About