In this paper we address the problem of automated three dimensional point cloud interpretation. This problem is important for various tasks from environment modeling to obstacle avoidance for autonomous robot navigation. In addition to locally extracted features, classifiers need to utilize contextual information in order to perform well. A popular approach to account for context is to utilize the Markov Random Field framework. One recent variant that has successfully been used for the problem considered is the Associative Markov Network (AMN). We extend the AMN model to learn directionality in the clique potentials, resulting in a new anisotropic model that can be efficiently learned using the subgradient method. We validate the proposed approach using data collected from different range sensors and show better performance against standard AMN and Support Vector Machine algorithms. 1.

Abstract — This paper introduces a probabilistic, two-stage classification framework for the semantic annotation of urban maps as provided by a mobile robot. During the first stage, local scene properties are considered using a probabilistic bagof-words classifier. The second stage incorporates contextual information across a given scene via a Markov Random Field (MRF). Our approach is driven by data from an onboard camera and 3D laser scanner and uses a combination of appearancebased and geometric features. By framing the classification exercise probabilistically we are able to execute an informationtheoretic bail-out policy when evaluating appearance-based classconditional likelihoods. This efficiency, combined with low order MRFs resulting from our two-stage approach, allows us to generate scene labels at speeds suitable for online deployment and use. We demonstrate and analyze the performance of our technique on data gathered over almost 17 km of track through a city. I.

Summary. This paper is concerned with assessing the quality of work-space maps. While there has been much work in recent years on building maps of field settings, little attention has been given to endowing a machine with introspective competencies which would allow assessing the reliability/plausibility of the representation. We classify regions in 3D point-cloud maps into two binary classes — “plausible ” or “suspicious”. In this paper we concentrate on the classification of urban maps and use a Conditional Random Fields to model the intrinsic qualities of planar patches and crucially, their relationship to each other. A bipartite labelling of the map is acquired via application of the Graph Cut algorithm. We present results using data gathered by a mobile robot equipped with a 3D laser range sensor while operating in a typical urban setting. 1

Abstract This paper presents algorithms for fast segmentation of 3D point clouds and subsequent classification of the obtained 3D segments. The method jointly determines the ground surface and segments individual objects in 3D, including overhanging structures. When compared to six other terrain modelling techniques, this approach has minimal error between the sensed data and the representation; and is fast (processing a Velodyne scan in approximately 2 seconds). Applications include improved alignment of successive scans by enabling operations in sections (Velodyne scans are aligned 7 % sharper compared to an approach using raw points) and more informed decision-making (paths move around overhangs). The use of segmentation to aid classification through 3D features, such as the Spin Image or the Spherical Harmonic Descriptor, is discussed and experimentally compared. Moreover, the segmentation facilitates a novel approach to 3D classification that bypasses feature extraction and directly compares 3D shapes via the ICP algorithm. This technique is shown to achieve accuracy on par with the best feature based classifier (92.1%) while being significantly faster and allowing a clearer understanding of the classifier’s behaviour. 1.1

Abstract — Segmenting range data into semantic categories has become a more and more active field of research in robotics. In this paper, we advocate to view this task as a problem of fast, large-scale retrieval. Intuitively, given a dataset of millions of labeled scan points and their neighborhoods, we simply search for similar points in the datasets and use the labels of the retrieved ones to predict the labels of a novel point using some local prediction model such as majority vote or logistic regression. However, actually carrying this out requires highly efficient ways of (1) storing millions of scan points in memory and (2) quickly finding similar scan points to a target scan point. In this paper, we propose to address both issues by employing Weiss et al.’s recent spectral hashing. It represents each item in a database by a compact binary code that is constructed so that similar items will have similar binary code words. In turn, similar neighbors have codes within a small Hamming distance of the code for the query. Then, we learn a logistic regression model locally over all points with the same binary code word. Our experiments on real world 3D scans show that the resulting approach, called spectrally hashed logistic regression, can be ultra fast at prediction time and outperforms state-of-the art approaches such as logistic regression and nearest neighbor. I.

Abstract — Truly versatile robots operating in the real world have to be able to learn about objects and their properties autonomously, that is, without being provided with carefully engineered training data. This paper presents an approach that allows a robot to discover object classes in three-dimensional range data in an unsupervised fashion and without a-priori knowledge about the observed objects. Our approach builds on Latent Dirichlet Allocation (LDA), a recently proposed probabilistic method for discovering topics in text documents. We discuss feature extraction, hypothesis generation, and statistical modeling of objects in 3D range data as well as the novel application of LDA to this domain. Our approach has been implemented and evaluated on real data of complex objects. Practical experiments demonstrate, that our approach is able to learn object class models autonomously that are consistent with the true classifications provided by a human. It furthermore outperforms unsupervised method such as hierarchical clustering that operate on a distance metric. I.

Abstract — In this paper, we investigate the problem of 3D object categorization of objects typically present in kitchen environments, from data acquired using a composite sensor. Our framework combines different sensing modalities and defines descriptive features in various spaces for the purpose of learning good object models. By fusing the 3D information acquired from a composite sensor that includes a color stereo camera, a timeof-flight (TOF) camera, and a thermal camera, we augment 3D depth data with color and temperature information which helps disambiguate the object categorization process. We make use of statistical relational learning methods (Markov Logic Networks and Bayesian Logic Networks) to capture complex interactions between the different feature spaces. To show the effectiveness of our approach, we analyze and validate the proposed system for the problem of recognizing objects in table settings scenarios. sensor is also present in the figure but it is not used for the purpose of the experiments presented in this paper. Each of the individual sensors cannot obtain enough information for the task of object categorization on its own. Stereo cameras, for example, cannot provide the necessary depth information for untextured planes, glasses are transparent and not easily visible with any camera, and thermal cameras can only provide partial information about the world when the temperature is different. However, combining the complementary perceptual evidence provided by the individual sensors gives us a much more valuable source of information. I.

Generating rich representations of environments can significantly improve the autonomy of mobile robotics. In this paper we introduce a novel approach to building object-type maps of outdoor environments. Our approach uses conditional random fields (CRF) to jointly classify laser returns in a 2D scan map into seven object types (car, wall, tree trunk, foliage, person, grass, and other). The spatial connectivity of the CRF model is determined via Delaunay triangulation of the laser map. Our model incorporates laser shape features, visual appearance features, structural information extracted from clusters of laser returns, and visual object detectors trained on image data sets available on the internet. The parameters of the CRF are trained from partially labeled laser and camera data collected by a car moving through an urban environment. Our approach achieves 91 % accuracy in classifying objects observed along a 3 kilometer trajectory.

Summary. This paper is concerned with assessing the quality of work-space maps. While there has been much work in recent years on building maps of field settings, little attention has been given to endowing a machine with introspective competencies which would allow assessing the reliability/plausibility of the representation. We classify regions in 3D point-cloud maps into two binary classes — “plausible ” or “suspicious”. In this paper we concentrate on the classification of urban maps and use a Conditional Random Fields to model the intrinsic qualities of planar patches and crucially, their relationship to each other. A bipartite labelling of the map is acquired via application of the Graph Cut algorithm. We present results using data gathered by a mobile robot equipped with a 3D laser range sensor while operating in a typical urban setting. 1