Centroid

About: Centroid is a research topic. Over the lifetime, 4110 publications have been published within this topic receiving 53637 citations. The topic is also known as: barycenter (geometry) & geometric center of a plane figure.

Centroid-Based Document Classification: Analysis and Experimental Results

Abstract: In this paper we present a simple linear-time centroid-based document classification algorithm, that despite its simplicity and robust performance, has not been extensively studied and analyzed. Our experiments show that this centroidbased classifier consistently and substantially outperforms other algorithms such as Naive Bayesian, k-nearest-neighbors, and C4.5, on a wide range of datasets. Our analysis shows that the similarity measure used by the centroid-based scheme allows it to classify a new document based on how closely its behavior matches the behavior of the documents belonging to different classes. This matching allows it to dynamically adjust for classes with different densities and accounts for dependencies between the terms in the different classes

Deep Hough Voting for 3D Object Detection in Point Clouds

Abstract: Current 3D object detection methods are heavily influenced by 2D detectors. In order to leverage architectures in 2D detectors, they often convert 3D point clouds to regular grids (i.e., to voxel grids or to bird's eye view images), or rely on detection in 2D images to propose 3D boxes. Few works have attempted to directly detect objects in point clouds. In this work, we return to first principles to construct a 3D detection pipeline for point cloud data and as generic as possible. However, due to the sparse nature of the data -- samples from 2D manifolds in 3D space -- we face a major challenge when directly predicting bounding box parameters from scene points: a 3D object centroid can be far from any surface point thus hard to regress accurately in one step. To address the challenge, we propose VoteNet, an end-to-end 3D object detection network based on a synergy of deep point set networks and Hough voting. Our model achieves state-of-the-art 3D detection on two large datasets of real 3D scans, ScanNet and SUN RGB-D with a simple design, compact model size and high efficiency. Remarkably, VoteNet outperforms previous methods by using purely geometric information without relying on color images.

Surface-marker cluster design criteria for 3-D bone movement reconstruction

Abstract: When three-dimensional (3-D) human or animal movement is recorded using a photogrammetric system, bone-embedded frame positions and orientations are estimated from reconstructed surface marker trajectories using either nonoptimal or optimal algorithms. The effectiveness of these mathematical procedures in accommodating for both photogrammetric errors and skin movement artifacts depends on the number of markers associated with a given bone as well as on the size and shape characteristics of the relevant cluster. One objective of this paper deals with the identification of marker cluster design criteria aimed at the minimization of error propagation from marker coordinates to bone-embedded frame position and orientation. Findings allow for the quantitative estimation of these errors for any given cluster configuration and suggest the following main design criteria. A cluster made up of four markers represents a good practical compromise. Planar clusters are acceptable, provided in quasi-isotropic distribution. The root mean square distance of the markers from their centroid should be greater than ten times the standard deviation of the marker position error. The second objective of this paper deals with the identification of the optimal cluster position and orientation on the limb aimed at the minimization of error propagation to anatomical landmark laboratory coordinates. Cluster position should be selected to minimize skin movement artifacts. The longest principal axis of the marker distribution should be oriented toward the relevant anatomical landmark position.