Multiview point cloud kernels for semisupervised learning [Lecture Notes]

In semisupervised learning (SSL), a predictive model is learn from a collection of labeled data and a typically much larger collection of unlabeled data. These paper presented a framework called multi-view point cloud regularization (MVPCR), which unifies and generalizes several semisupervised kernel methods that are based on data-dependent regularization in reproducing kernel Hilbert spaces (RKHSs). Special cases of MVPCR include coregularized least squares (CoRLS), manifold regularization (MR), and graph-based SSL. An accompanying theorem shows how to reduce any MVPCR problem to standard supervised learning with a new multi-view kernel.

Lost in translation (and rotation) : rapid extrinsic calibration for 2D and 3D LIDARs

This paper describes a novel method for determining the extrinsic calibration parameters between 2D and 3D LIDAR sensors with respect to a vehicle base frame. To recover the calibration parameters we attempt to optimize the quality of a 3D point cloud produced by the vehicle as it traverses an unknown, unmodified environment. The point cloud quality metric is derived from Rényi Quadratic Entropy and quantifies the compactness of the point distribution using only a single tuning parameter. We also present a fast approximate method to reduce the computational requirements of the entropy evaluation, allowing unsupervised calibration in vast environments with millions of points. The algorithm is analyzed using real world data gathered in many locations, showing robust calibration performance and substantial speed improvements from the approximations.

Advantages of exploiting projection structure for segmenting dense 3D point clouds

Timely and comprehensive scene segmentation is often a critical step for many high level mobile robotic tasks. This paper examines a projected area based neighbourhood lookup approach with the motivation towards faster unsupervised segmentation of dense 3D point clouds. The proposed algorithm exploits the projection geometry of a depth camera to find nearest neighbours which is time independent of the input data size. Points near depth discontinuations are also detected to reinforce object boundaries in the clustering process. The search method presented is evaluated using both indoor and outdoor dense depth images and demonstrates significant improvements in speed and precision compared to the commonly used Fast library for approximate nearest neighbour (FLANN) [Muja and Lowe, 2009].

Efficiently capturing large, complex cultural heritage sites with a handheld mobile 3D laser mapping system

Accurate three-dimensional representations of cultural heritage sites are highly valuable for scientific study, conservation, and educational purposes. In addition to their use for archival purposes, 3D models enable efficient and precise measurement of relevant natural and architectural features. Many cultural heritage sites are large and complex, consisting of multiple structures spatially distributed over tens of thousands of square metres. The process of effectively digitising such geometrically complex locations requires measurements to be acquired from a variety of viewpoints. While several technologies exist for capturing the 3D structure of objects and environments, none are ideally suited to complex, large-scale sites, mainly due to their limited coverage or acquisition efficiency. We explore the use of a recently developed handheld mobile mapping system called Zebedee in cultural heritage applications. The Zebedee system is capable of efficiently mapping an environment in three dimensions by continually acquiring data as an operator holding the device traverses through the site. The system was deployed at the former Peel Island Lazaret, a culturally significant site in Queensland, Australia, consisting of dozens of buildings of various sizes spread across an area of approximately 400 × 250 m. With the Zebedee system, the site was scanned in half a day, and a detailed 3D point cloud model (with over 520 million points) was generated from the 3.6 hours of acquired data in 2.6 hours. We present results demonstrating that Zebedee was able to accurately capture both site context and building detail comparable in accuracy to manual measurement techniques, and at a greatly increased level of efficiency and scope. The scan allowed us to record derelict buildings that previously could not be measured because of the scale and complexity of the site. The resulting 3D model captures both interior and exterior features of buildings, including structure, materials, and the contents of rooms.

Separable spatiotemporal priors for convex reconstruction of time-varying 3D point clouds

Reconstructing 3D motion data is highly under-constrained due to several common sources of data loss during measurement, such as projection, occlusion, or miscorrespondence. We present a statistical model of 3D motion data, based on the Kronecker structure of the spatiotemporal covariance of natural motion, as a prior on 3D motion. This prior is expressed as a matrix normal distribution, composed of separable and compact row and column covariances. We relate the marginals of the distribution to the shape, trajectory, and shape-trajectory models of prior art. When the marginal shape distribution is not available from training data, we show how placing a hierarchical prior over shapes results in a convex MAP solution in terms of the trace-norm. The matrix normal distribution, fit to a single sequence, outperforms state-of-the-art methods at reconstructing 3D motion data in the presence of significant data loss, while providing covariance estimates of the imputed points.

Vision-only autonomous navigation using topometric maps

This paper presents a mapping and navigation system for a mobile robot, which uses vision as its sole sensor modality. The system enables the robot to navigate autonomously, plan paths and avoid obstacles using a vision based topometric map of its environment. The map consists of a globally-consistent pose-graph with a local 3D point cloud attached to each of its nodes. These point clouds are used for direction independent loop closure and to dynamically generate 2D metric maps for locally optimal path planning. Using this locally semi-continuous metric space, the robot performs shortest path planning instead of following the nodes of the graph --- as is done with most other vision-only navigation approaches. The system exploits the local accuracy of visual odometry in creating local metric maps, and uses pose graph SLAM, visual appearance-based place recognition and point clouds registration to create the topometric map. The ability of the framework to sustain vision-only navigation is validated experimentally, and the system is provided as open-source software.

3D anthropometry : from the ergonomist's perspective

Since the availability of 3D full body scanners and the associated software systems for operations with large point clouds, 3D anthropometry has been marketed as a breakthrough and milestone in ergonomic design. The assumptions made by the representatives of the 3D paradigm need to be critically reviewed though. 3D anthropometry has advantages as well as shortfalls, which need to be carefully considered. While it is apparent that the measurement of a full body point cloud allows for easier storage of raw data and improves quality control, the difficulties in calculation of standardized measurements from the point cloud are widely underestimated. Early studies that made use of 3D point clouds to derive anthropometric dimensions have shown unacceptable deviations from the standardized results measured manually. While 3D human point clouds provide a valuable tool to replicate specific single persons for further virtual studies, or personalize garment, their use in ergonomic design must be critically assessed. Ergonomic, volumetric problems are defined by their 2-dimensional boundary or one dimensional sections. A 1D/2D approach is therefore sufficient to solve an ergonomic design problem. As a consequence, all modern 3D human manikins are defined by the underlying anthropometric girths (2D) and lengths/widths (1D), which can be measured efficiently using manual techniques. Traditionally, Ergonomists have taken a statistical approach to design for generalized percentiles of the population rather than for a single user. The underlying method is based on the distribution function of meaningful single and two-dimensional anthropometric variables. Compared to these variables, the distribution of human volume has no ergonomic relevance. On the other hand, if volume is to be seen as a two-dimensional integral or distribution function of length and girth, the calculation of combined percentiles – a common ergonomic requirement - is undefined. Consequently, we suggest to critically review the cost and use of 3D anthropometry. We also recommend making proper use of widely available single and 2-dimensional anthropometric data in ergonomic design.

Three dimensional digitisation of plant leaves

Realistic plant models are important for leaf area and plant volume estimation, reconstruction of growth canopies, structure generation of the plant, reconstruction of leaf surfaces and agrichemical spray droplet modelling. This article investigates several different scanning devices for obtaining a three dimensional digitisation of plant leaves with a point cloud resolution of 200-500μm. The devices tested were a Roland mdx-20, Microsoft Kinect, Roland lpx-250, Picoscan and Artec S. The applicability of each of these devices for scanning plant leaves is discussed. The most suitable tested digitisation device for scanning plant leaves is the Artec S scanner.

An application of fuzzy DL-Based semantic perception to soil container classification

Semantic perception and object labeling are key requirements for robots interacting with objects on a higher level. Symbolic annotation of objects allows the usage of planning algorithms for object interaction, for instance in a typical fetchand-carry scenario. In current research, perception is usually based on 3D scene reconstruction and geometric model matching, where trained features are matched with a 3D sample point cloud. In this work we propose a semantic perception method which is based on spatio-semantic features. These features are defined in a natural, symbolic way, such as geometry and spatial relation. In contrast to point-based model matching methods, a spatial ontology is used where objects are rather described how they "look like", similar to how a human would described unknown objects to another person. A fuzzy based reasoning approach matches perceivable features with a spatial ontology of the objects. The approach provides a method which is able to deal with senor noise and occlusions. Another advantage is that no training phase is needed in order to learn object features. The use-case of the proposed method is the detection of soil sample containers in an outdoor environment which have to be collected by a mobile robot. The approach is verified using real world experiments.

Classification of Airborne LIDAR Intensity Data Using Statistical Analysis and Hough Transform with Application to Power Line Corridors

Light Detection and Ranging (LIDAR) has great potential to assist vegetation management in power line corridors by providing more accurate geometric information of the power line assets and vegetation along the corridors. However, the development of algorithms for the automatic processing of LIDAR point cloud data, in particular for feature extraction and classification of raw point cloud data, is in still in its infancy. In this paper, we take advantage of LIDAR intensity and try to classify ground and non-ground points by statistically analyzing the skewness and kurtosis of the intensity data. Moreover, the Hough transform is employed to detected power lines from the filtered object points. The experimental results show the effectiveness of our methods and indicate that better results were obtained by using LIDAR intensity data than elevation data.

Multi-level information fusion for environment aware robotic navigation

This thesis develops the hardware and software framework for an integrated navigation system. Dynamic data fusion algorithms are used to develop a system with a high level of resistance to the typical problems that affect standard navigation systems.

BDNF gene effects on brain circuitry replicated in 455 twins

Brain-derived neurotrophic factor (BDNF) plays a key role in learning and memory, but its effects on the fiber architecture of the living brain are unknown. We genotyped 455 healthy adult twins and their non-twin siblings (188 males/267 females; age: 23.7 ± 2.1. years, mean ± SD) and scanned them with high angular resolution diffusion tensor imaging (DTI), to assess how the BDNF Val66Met polymorphism affects white matter microstructure. By applying genetic association analysis to every 3D point in the brain images, we found that the Val-BDNF genetic variant was associated with lower white matter integrity in the splenium of the corpus callosum, left optic radiation, inferior fronto-occipital fasciculus, and superior corona radiata. Normal BDNF variation influenced the association between subjects' performance intellectual ability (as measured by Object Assembly subtest) and fiber integrity (as measured by fractional anisotropy; FA) in the callosal splenium, and pons. BDNF gene may affect the intellectual performance by modulating the white matter development. This combination of genetic association analysis and large-scale diffusion imaging directly relates a specific gene to the fiber microstructure of the living brain and to human intelligence.

LiDAR Augmented Optimal Path Planning for Unmanned Aerial Vehicle during Remote Sensing at Low Altitude and Network Sensor Data Acquisition

This technical report describes a Light Detection and Ranging (LiDAR) augmented optimal path planning at low level flight methodology for remote sensing and sampling Unmanned Aerial Vehicles (UAV). The UAV is used to perform remote air sampling and data acquisition from a network of sensors on the ground. The data that contains information on the terrain is in the form of a 3D point clouds maps is processed by the algorithms to find an optimal path. The results show that the method and algorithm are able to use the LiDAR data to avoid obstacles when planning a path from a start to a target point. The report compares the performance of the method as the resolution of the LIDAR map is increased and when a Digital Elevation Model (DEM) is included. From a practical point of view, the optimal path plan is loaded and works seemingly with the UAV ground station and also shows the UAV ground station software augmented with more accurate LIDAR data.

GPX - Gardens Point XML Information Retrieval at INEX 2004

Traditional information retrieval (IR) systems respond to user queries with ranked lists of relevant documents. The separation of content and structure in XML documents allows individual XML elements to be selected in isolation. Thus, users expect XML-IR systems to return highly relevant results that are more precise than entire documents. In this paper we describe the implementation of a search engine for XML document collections. The system is keyword based and is built upon an XML inverted file system. We describe the approach that was adopted to meet the requirements of Content Only (CO) and Vague Content and Structure (VCAS) queries in INEX 2004.

Measuring Visual Consistency in 3D Rendering Systems

One of the major challenges facing a present day game development company is the removal of bugs from such complex virtual environments. This work presents an approach for measuring the correctness of synthetic scenes generated by a rendering system of a 3D application, such as a computer game. Our approach builds a database of labelled point clouds representing the spatiotemporal colour distribution for the objects present in a sequence of bug-free frames. This is done by converting the position that the pixels take over time into the 3D equivalent points with associated colours. Once the space of labelled points is built, each new image produced from the same game by any rendering system can be analysed by measuring its visual inconsistency in terms of distance from the database. Objects within the scene can be relocated (manually or by the application engine); yet the algorithm is able to perform the image analysis in terms of the 3D structure and colour distribution of samples on the surface of the object. We applied our framework to the publicly available game RacingGame developed for Microsoft(R) Xna(R). Preliminary results show how this approach can be used to detect a variety of visual artifacts generated by the rendering system in a professional quality game engine.