Superpixel Benchmark
Superpixel benchmark, tools and algorithms.
Class Hierarchy
This inheritance list is sorted roughly, but not completely, alphabetically:
[detail level 12]
 CDepthTools::CameraRepresents basic intrinsic parameters used to project pixels into a point cloud
 CCCS_OpenCVWrapper for running CCS using OpenCV images
 CCIS_OpenCVWrapper for running CIS with OpenCV images
 CConnectedComponentsEfficient (multi-label) connected components algorithm
 CCRS_OpenCVWrapper for running CRS on OpenCV images
 CDASP_OpenCVWrapper for running DASP on OpenCV images
 CDepthToolsTools for using depth information
 CERGC_OpenCVWrapper for running ERGC on OpenCV images
 CERS_OpenCVWrapper for running ERS on OpenCV images
 CETPS_OpenCVWrapper for running ETPS on OpenCV images
 CEvaluationProvides measures to evaluate (over-) segmentations
 CEvaluationSummary::EvaluationMetricsAvailable metrics, see evaluation.h
 CEvaluationSummary::EvaluationStatisticsAvailable statistics that may be computed for each metric
 CEvaluationSummaryGiven a directory of superpixel segmentations and a directory of ground truth segmentations, this class is used to generate a CSV file of statistics of different metrics
 CFH_OpenCVWrapper for running FH on OpenCV images
 CGraphSegmentationImplementation of graph based image segmentation as described in the paper by Felzenswalb and Huttenlocher
 CGraphSegmentationDistanceInterface to be implemented by a concerete distance. The distance defines how the weights between nodes in the image graph are computed. See the paper by Felzenswalb and Huttenlocher for details. Essentially, derived classes only need to overwrite the () operator
 CGraphSegmentationEuclideanRGBEuclidean RGB distance
 CGraphSegmentationManhattenRGBManhatten (i.e. L1) distance
 CGraphSegmentationMagicThe magic part of the graph segmentation, i.e. s given two nodes decide whether to add an edge between them (i.e. merge the corresponding segments). See the paper by Felzenswalb and Huttenlocher for details
 CGraphSegmentationMagicThreshold
 CImageEdgeRepresents an edge between two pixels in an image. Each edge is characterized by a weight and the adjacent nodes
 CImageEdgeSorterClass for sorting edges according to weight
 CImageGraphRepresents an image graph, consisting of one node per pixel which are 4-connected
 CImageNodeRepresents a pixel in a video. Each pixel is represented by its color which is needed to compute the weights between pixels
 CIOUtilI/O utilities
 CLSC_OpenCVWrapper for running LSC on OpenCV images
 CMSS_OpenCVWrapper for running MSS on OpenCV images
 CParameterOptimizationToolTool to guide parameter optimization using grid search
 CPB_OpenCVWrapper for running PB on OpenCV images
 CRobustnessToolTool evaluating the robustness for different filters/enhancement/transformations
 CRobustnessToolDriverDriver for different filters/enhancements/transformations
 CBilateralFilterDriverBilateral filter driver
 CBlurDriverBlur driver
 CGaussianBlurDriverGaussian blur filter
 CGaussianNoiseDriverAdditive Gaussian noise and Gaussian sampling error driver
 CMedianBlurDriverMedian blur driver
 CMotionBlurDriverMotion blur driver
 CPoissonNoiseDriverPoisson noise driver
 CRotationDriverRotation driver
 CSaltAndPepperNoiseDriverSalt and pepper noise driver
 CShearDriverVertical and horizontal shear driver
 CTranslationDriverTranslation driver
 Cscost
 CSLIC_OpenCVWrapper for running SLIC on OpenCV images
 CSuperpixelToolsSuperpixel utilities
 CEvaluationSummary::SuperpixelVisualizationsAvailable superpixel visualizations, see visualization.h
 CTransformationUtility class for image and ground truth transofmrations
 CVC_OpenCVWrapper for running VC on OpenCV images
 CVCCS_OpenCV_PCLWrapper for running PCL on OpenCV images given the point cloud as OpenCV image
 CVisualizationSome basic visualizations of superpixel segmentations
 CVLSLIC_OpenCVWrapper for running vlSLIC on OpenCV images