The Experts below are selected from a list of 318 Experts worldwide ranked by ideXlab platform
Gamini Dissanayake - One of the best experts on this subject based on the ideXlab platform.
-
Efficient Monocular SLAM using Sparse Information filters
2010 Fifth International Conference on Information and Automation for Sustainability, 2010Co-Authors: Zhan Wang, Gamini DissanayakeAbstract:A new method for efficiently mapping three dimensional environments from a platform carrying a single calibrated camera, and simultaneously localizing the platform within this map is presented in this paper. This is the Monocular SLAM problem in robotics, which is equivalent to the problem of extracting Structure from Motion (SFM) in computer vision. A novel formulation of Monocular SLAM which exploits recent results from multi-view geometry to partition the feature location measurements extracted from images into providing estimates of environment representation and platform motion is developed. Proposed formulation allows rich geometric Information from a large set of features extracted from images to be maximally incorporated during the estimation process, without a corresponding increase in the computational cost, resulting in more accurate estimates. A Sparse Extended Information Filter (EIF) which fully exploits the Sparse structure of the problem is used to generate camera pose and feature location estimates. Experimental results are provided to verify the algorithm.
-
ICARCV - Map-aided 6-DOF relative pose estimation for monocular SLAM using Sparse Information filters
2010 11th International Conference on Control Automation Robotics & Vision, 2010Co-Authors: Zhan Wang, Gamini DissanayakeAbstract:This paper addresses the problem of mapping three dimensional environments from a sequence of images taken by a calibrated camera, and simultaneously generating the camera motion trajectory. This is the Monocular SLAM problem in robotics, and is akin to the Structure from Motion (SFM) problem in computer vision. We present a novel map-aided 6-DOF relative pose estimation method based on a new formulation of the Monocular SLAM that is able to provide better initial estimates of new camera poses than the simple triangulation traditionally used in this context. The '6-DOF' means relative to the map which itself is up to an unobservable scale. The proposed pose estimator also allows more effective outlier rejection in matching features present in the map and features extracted from two consecutive images. Our Monocular SLAM algorithm is able to deal with arbitrary camera motion, making the smooth motion assumption, which is required by the typically used constant velocity model, unnecessary. In the new Monocular SLAM formulation, the measurements of extracted features from images are partitioned into those used for the estimation of the environment and those used for estimating the camera motion. The new formulation enables the current map estimate to aid achieving the full 6-DOF relative pose estimation up to the mapping scale while maximally exploiting the geometry Information in images. Experiment results are provided to verify the proposed algorithm.
-
SIMULTANEOUS LOCALIZATION AND MAPPING: Exactly Sparse Information Filters
New Frontiers in Robotics, 2005Co-Authors: Zhan Wang, Shoudong Huang, Gamini DissanayakeAbstract:Simultaneous localization and mapping (SLAM) is a process where an autonomous vehicle builds a map of an unknown environment while concurrently generating an estimate for its location. This book is concerned with computationally efficient solutions to the large scale SLAM problems using exactly Sparse Extended Information Filters (EIF).The invaluable book also provides a comprehensive theoretical analysis of the properties of the Information matrix in EIF-based algorithms for SLAM. Three exactly Sparse Information filters for SLAM are described in detail, together with two efficient and exact methods for recovering the state vector and the covariance matrix. Proposed algorithms are extensively evaluated both in simulation and through experiments.
-
FSR - Tradeoffs in SLAM with Sparse Information filters
Springer Tracts in Advanced Robotics, 1Co-Authors: Zhan Wang, Shoudong Huang, Gamini DissanayakeAbstract:Designing filters exploiting the Sparseness of the Information matrix for efficiently solving the simultaneous localization and mapping (SLAM) problem has attracted significant attention during the recent past. The main contribution of this paper is a review of the various Sparse Information filters proposed in the literature to date, in particular, the compromises used to achieve Sparseness. Two of the most recent algorithms that the authors have implemented, Exactly Sparse Extended Information Filter (ESEIF) by Walter et al. [5] and the D-SLAM by Wang et al. [6] are discussed and analyzed in detail. It is proposed that this analysis can stimulate developing a framework suitable for evaluating the relative merits of SLAM algorithms.
Zhan Wang - One of the best experts on this subject based on the ideXlab platform.
-
Efficient Monocular SLAM using Sparse Information filters
2010 Fifth International Conference on Information and Automation for Sustainability, 2010Co-Authors: Zhan Wang, Gamini DissanayakeAbstract:A new method for efficiently mapping three dimensional environments from a platform carrying a single calibrated camera, and simultaneously localizing the platform within this map is presented in this paper. This is the Monocular SLAM problem in robotics, which is equivalent to the problem of extracting Structure from Motion (SFM) in computer vision. A novel formulation of Monocular SLAM which exploits recent results from multi-view geometry to partition the feature location measurements extracted from images into providing estimates of environment representation and platform motion is developed. Proposed formulation allows rich geometric Information from a large set of features extracted from images to be maximally incorporated during the estimation process, without a corresponding increase in the computational cost, resulting in more accurate estimates. A Sparse Extended Information Filter (EIF) which fully exploits the Sparse structure of the problem is used to generate camera pose and feature location estimates. Experimental results are provided to verify the algorithm.
-
ICARCV - Map-aided 6-DOF relative pose estimation for monocular SLAM using Sparse Information filters
2010 11th International Conference on Control Automation Robotics & Vision, 2010Co-Authors: Zhan Wang, Gamini DissanayakeAbstract:This paper addresses the problem of mapping three dimensional environments from a sequence of images taken by a calibrated camera, and simultaneously generating the camera motion trajectory. This is the Monocular SLAM problem in robotics, and is akin to the Structure from Motion (SFM) problem in computer vision. We present a novel map-aided 6-DOF relative pose estimation method based on a new formulation of the Monocular SLAM that is able to provide better initial estimates of new camera poses than the simple triangulation traditionally used in this context. The '6-DOF' means relative to the map which itself is up to an unobservable scale. The proposed pose estimator also allows more effective outlier rejection in matching features present in the map and features extracted from two consecutive images. Our Monocular SLAM algorithm is able to deal with arbitrary camera motion, making the smooth motion assumption, which is required by the typically used constant velocity model, unnecessary. In the new Monocular SLAM formulation, the measurements of extracted features from images are partitioned into those used for the estimation of the environment and those used for estimating the camera motion. The new formulation enables the current map estimate to aid achieving the full 6-DOF relative pose estimation up to the mapping scale while maximally exploiting the geometry Information in images. Experiment results are provided to verify the proposed algorithm.
-
Exactly Sparse Information filters for simultaneous localization and mapping
2007Co-Authors: Zhan WangAbstract:NO FULL TEXT AVAILABLE. Access is restricted indefinitely. – – This thesis is concerned with computationally efficient solutions to the simultaneous localization and mapping (SLAM) problem. The setting for the SLAM problem is that of a robot with a known kinematic model, equipped with on-board sensors, moving through an environment consisting of a population of features. The objective of the SLAM problem is to estimate the position and orientation of the robot together with the locations of all the features. Extended Kalman Filter (EKF) based SLAM solutions widely discussed in the literature require the maintenance of a large and dense covariance matrix. Recently, Extended Information Filter (ElF) based SLAM solutions have attracted significant attention due to the recognition that the associated Information matrix can be made Sparse. However, existing algorithms for ElF-based SLAM have a number of disadvantages, such as estimator inconsistency, a long state vector or Information loss, as a consequence of the strategies used for achieving the Sparseness of the Information matrix. Furthermore, some important practical issues such as the efficient recovery of the state estimate and the associated covariance matrix need further work. The contributions of this thesis include three new exactly Sparse Information filters for SLAM: one is achieved by decoupling the localization and mapping processes in SLAM; the other two are aimed at SLAM in large environments through joining many small scale local maps. In the first algorithm, D-SLAM, it is shown that SLAM can be performed in a decoupled manner in which the localization and mapping are two separate yet concurrent processes. This formulation of SLAM results in a new and natural way to achieve the Sparse Information matrix without any approximation. In the second algorithm, the relative Information present in each local map is first extracted, and then used to build a global map based on the D-SLAM framework. Both these algorithms, while computationally efficient, incur some Information loss. The third algorithm that modifies the global map state vector by incorporating robot start and end poses of each local map, completely avoids the Information loss while maintaining the Sparseness of the Information matrix and associated computational advantages. Two efficient methods for recovering the state estimate and the associated covariance matrix from the output of the ElF are also proposed. These methods exploit the gradual evolution of the SLAM Information matrix, and allow the ElF-based SLAM algorithms proposed in this thesis to be implemented at a computational cost that is linearly proportional to the size of the map.
-
SIMULTANEOUS LOCALIZATION AND MAPPING: Exactly Sparse Information Filters
New Frontiers in Robotics, 2005Co-Authors: Zhan Wang, Shoudong Huang, Gamini DissanayakeAbstract:Simultaneous localization and mapping (SLAM) is a process where an autonomous vehicle builds a map of an unknown environment while concurrently generating an estimate for its location. This book is concerned with computationally efficient solutions to the large scale SLAM problems using exactly Sparse Extended Information Filters (EIF).The invaluable book also provides a comprehensive theoretical analysis of the properties of the Information matrix in EIF-based algorithms for SLAM. Three exactly Sparse Information filters for SLAM are described in detail, together with two efficient and exact methods for recovering the state vector and the covariance matrix. Proposed algorithms are extensively evaluated both in simulation and through experiments.
-
FSR - Tradeoffs in SLAM with Sparse Information filters
Springer Tracts in Advanced Robotics, 1Co-Authors: Zhan Wang, Shoudong Huang, Gamini DissanayakeAbstract:Designing filters exploiting the Sparseness of the Information matrix for efficiently solving the simultaneous localization and mapping (SLAM) problem has attracted significant attention during the recent past. The main contribution of this paper is a review of the various Sparse Information filters proposed in the literature to date, in particular, the compromises used to achieve Sparseness. Two of the most recent algorithms that the authors have implemented, Exactly Sparse Extended Information Filter (ESEIF) by Walter et al. [5] and the D-SLAM by Wang et al. [6] are discussed and analyzed in detail. It is proposed that this analysis can stimulate developing a framework suitable for evaluating the relative merits of SLAM algorithms.
G. D. Davies - One of the best experts on this subject based on the ideXlab platform.
-
length weight relationships of the invasive topmouth gudgeon pseudorasbora parva in ten lakes in the uk
Journal of Applied Ichthyology, 2007Co-Authors: J. R. Britton, G. D. DaviesAbstract:Summary The topmouth gudgeon (Pseudorasbora parva) is a species that is becoming invasive in UK waters. Few biological studies have been completed on the species in this extended range, resulting in Sparse Information in some aspects of their invasion biology. Regarding length–weight relationships, this is remedied in this study by providing data for 10 lake populations in the UK. Using fish length range 2.5–11.8 cm, values of a and b in the length–weight equation are provided. Values of b ranged between 2.76 and 3.32. Across all lakes, the relationship was W = 0.011L3.03, with weight in grams and length (fork) in cm.
-
Length–weight relationships of the invasive topmouth gudgeon (Pseudorasbora parva) in ten lakes in the UK
Journal of Applied Ichthyology, 2007Co-Authors: J. R. Britton, G. D. DaviesAbstract:Summary The topmouth gudgeon (Pseudorasbora parva) is a species that is becoming invasive in UK waters. Few biological studies have been completed on the species in this extended range, resulting in Sparse Information in some aspects of their invasion biology. Regarding length–weight relationships, this is remedied in this study by providing data for 10 lake populations in the UK. Using fish length range 2.5–11.8 cm, values of a and b in the length–weight equation are provided. Values of b ranged between 2.76 and 3.32. Across all lakes, the relationship was W = 0.011L3.03, with weight in grams and length (fork) in cm.
Oren Etzioni - One of the best experts on this subject based on the ideXlab platform.
-
Sparse Information extraction unsupervised language models to the rescue
Meeting of the Association for Computational Linguistics, 2007Co-Authors: Doug Downey, Stefan Schoenmackers, Oren EtzioniAbstract:Even in a massive corpus such as the Web, a substantial fraction of extractions appear infrequently. This paper shows how to assess the correctness of Sparse extractions by utilizing unsupervised language models. The REALM system, which combines HMMbased and n-gram-based language models, ranks candidate extractions by the likelihood that they are correct. Our experiments show that REALM reduces extraction error by 39%, on average, when compared with previous work. Because REALM pre-computes language models based on its corpus and does not require any hand-tagged seeds, it is far more scalable than approaches that learn models for each individual relation from handtagged data. Thus, REALM is ideally suited for open Information extraction where the relations of interest are not specified in advance and their number is potentially vast.
-
ACL - Sparse Information Extraction: Unsupervised Language Models to the Rescue
2007Co-Authors: Doug Downey, Stefan Schoenmackers, Oren EtzioniAbstract:Even in a massive corpus such as the Web, a substantial fraction of extractions appear infrequently. This paper shows how to assess the correctness of Sparse extractions by utilizing unsupervised language models. The REALM system, which combines HMMbased and n-gram-based language models, ranks candidate extractions by the likelihood that they are correct. Our experiments show that REALM reduces extraction error by 39%, on average, when compared with previous work. Because REALM pre-computes language models based on its corpus and does not require any hand-tagged seeds, it is far more scalable than approaches that learn models for each individual relation from handtagged data. Thus, REALM is ideally suited for open Information extraction where the relations of interest are not specified in advance and their number is potentially vast.
Eveline Baumgart-vogt - One of the best experts on this subject based on the ideXlab platform.
-
Peroxisomes in Human and Mouse Testis: Differential Expression of Peroxisomal Proteins in Germ Cells and Distinct Somatic Cell Types of the Testis
Biology of Reproduction, 2007Co-Authors: Anca Nenicu, Georg H. Lüers, Werner J. Kovacs, Martin Bergmann, Eveline Baumgart-vogtAbstract:Abstract The vital importance of peroxisomal metabolism for regular function of the testis is stressed by the severe spermatogenesis defects induced by peroxisomal dysfunction. However, only Sparse Information is available on the role and enzyme composition of this organelle in distinct cell types of the testis. In the present study, we characterized the peroxisomal compartment in human and mouse testis in primary cultures of murine somatic cells (Sertoli, peritubular myoid, and Leydig cells) and in GFP-PTS1 transgenic mice with a variety of morphological and biochemical techniques. Formerly, peroxisomes were thought to be absent in late stages of spermatogenesis. However, our results obtained by detection of different peroxisomal marker proteins show the presence of these organelles in most cell types in the testis, except for mature spermatozoa. Furthermore, we demonstrate a strong heterogeneity of peroxisomal protein content in various cell types of the human and mouse testis and show marked difference...
