Titlebook: Progress in Image Analysis and Processing, ICIAP 2013; Naples, Italy, Septe Alfredo Petrosino Conference proceedings 2013 Springer-Verlag G

Inflamed

Integrating Color Sampling into Depth Based Bilayer Segmentation,a novel approach for bi-layer segmentation, whose main contribution is given by the integration of a color classifier based on color sampling, within a depth-based segmentation framework. We have run tests on datasets available online and the outcoming results pointed out the effectiveness of this a

PATRI

急急忙忙

Deeply Optimized Hough Transform: Application to Action Segmentation, extraction or human action detection. However, these methods are known to generate false positives. To handle this issue, several works like . (.) or . (.) have reported significant performance improvements by adding discriminative parameters to the generative ones introduced by .. In this paper, w

邪恶的你

Layout-Based Document-Retrieval System by Radon Transform Using Dynamic Time Warping,y designed for commercial form retrieval. The system first uses a technique based on mathematical morphology to extract grid-based structural components from the document image. Successively, Radon Transform is used for document layout description. A document matching technique based on dynamic time

ELUDE

Evaluation of Low-Level Image Representations for Illumination-Insensitive Recognition of Texturelescriminative power of various low-level image features for a pixelwise representation of the underlying surface characteristics of the object. For this purpose, a new dataset with rendered images of 3D models is used which allows to directly compare the influences of texture and material properties

带伤害

Concrete

jet-lag

Nonporous

Approximating the Skeleton for Fine-to-Coarse Shape Representation,adius of the associated ball, i.e., with its distance from the complement of the object. Polygonal approximation is accomplished on all skeleton branches represented in a 4D space, where the coordinates are the (.,.,.) coordinates plus the radius . associated with each skeleton voxel. In this way, s

Torrid