In this paper we present a surface denoising technique using the vertex-centered finite volume method in conjunction with the mesh covariance fractional anisotropy. The approach is computationally fast and able to effectively remove undesirable noise while preserving prominent geometric features of a 3D mesh surface such as curved surface regions, sharp edges, and fine details. Extensive experimental results on various 3D models demonstrate the effectiveness of the proposed iterative algorithm, which yields satisfactory output results in just one single iteration.
{"title":"Fast Surface Denoising Using Finite Volumes of the Dual Mesh","authors":"Anis Kacem, A. Hamza","doi":"10.1109/CRV.2012.17","DOIUrl":"https://doi.org/10.1109/CRV.2012.17","url":null,"abstract":"In this paper we present a surface denoising technique using the vertex-centered finite volume method in conjunction with the mesh covariance fractional anisotropy. The approach is computationally fast and able to effectively remove undesirable noise while preserving prominent geometric features of a 3D mesh surface such as curved surface regions, sharp edges, and fine details. Extensive experimental results on various 3D models demonstrate the effectiveness of the proposed iterative algorithm, which yields satisfactory output results in just one single iteration.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132144671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We present a novel approach to the guidance of an autonomous underwater vehicle (AUV) along a trained route. The introduced system employs a topological route representation based on storing a sequence of side-scan sonar images captured along the route. When in following mode, image registration techniques provide the vehicle with a real-time estimate of the direction of its displacement relative to the trained route. This simplified approach to navigation sidesteps the problems inherent with maintaining a vehicle pose estimate within a global reference system, thereby allowing the vehicle to traverse a trained route without resorting to external navigation aides (e.g. GPS). Simulation results are provided which validate the proof of concept for our approach.
{"title":"Towards AUV Route Following Using Qualitative Navigation","authors":"P. Vandrish, A. Vardy, P. King","doi":"10.1109/CRV.2012.63","DOIUrl":"https://doi.org/10.1109/CRV.2012.63","url":null,"abstract":"We present a novel approach to the guidance of an autonomous underwater vehicle (AUV) along a trained route. The introduced system employs a topological route representation based on storing a sequence of side-scan sonar images captured along the route. When in following mode, image registration techniques provide the vehicle with a real-time estimate of the direction of its displacement relative to the trained route. This simplified approach to navigation sidesteps the problems inherent with maintaining a vehicle pose estimate within a global reference system, thereby allowing the vehicle to traverse a trained route without resorting to external navigation aides (e.g. GPS). Simulation results are provided which validate the proof of concept for our approach.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126082370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Mehta, P. Barooah, W. Dixon, E. Pasiliao, J. Curtis
In this paper, a robust pose (i.e., position and orientation) estimation algorithm using two-views captured by a calibrated monocular camera is presented. A collection of pose hypotheses is obtained when more than the minimum number of feature points required to uniquely identify a pose are available in both the images. The pose hypotheses - unit quaternion and unit translation vectors - lie on the S3 and S2 manifolds in the Euclidean 4-space and 3-space, respectively. Probability density function (pdf) of the rotation and translation pose hypotheses is evaluated by gridding the unit spheres where a robust, coarse pose estimate is identified at the mode of the pdf. Further, a "refining" pdf of the geodesic distance from the coarse pose estimate is constructed for the hypotheses within a grid containing the coarse estimate. A refined pose estimate is obtained by averaging the low-noise hypotheses in the neighbourhood of the mode of refining pdf. Pose estimation results of the proposed method are compared with RANSAC and nonlinear mean-shift (NMS) algorithms using the Oxford Corridor sequence and the robustness to feature outliers, image noise rejection, and scalability to number of features is analyzed using the synthetic data experiments. Processing time comparison with the RANSAC and NMS algorithms indicate that the deterministic time requirement of the proposed and NMS algorithms is amenable to a variety of visual servo control applications.
{"title":"PEGUS: An Image-Based Robust Pose Estimation Method","authors":"S. Mehta, P. Barooah, W. Dixon, E. Pasiliao, J. Curtis","doi":"10.1109/CRV.2012.18","DOIUrl":"https://doi.org/10.1109/CRV.2012.18","url":null,"abstract":"In this paper, a robust pose (i.e., position and orientation) estimation algorithm using two-views captured by a calibrated monocular camera is presented. A collection of pose hypotheses is obtained when more than the minimum number of feature points required to uniquely identify a pose are available in both the images. The pose hypotheses - unit quaternion and unit translation vectors - lie on the S3 and S2 manifolds in the Euclidean 4-space and 3-space, respectively. Probability density function (pdf) of the rotation and translation pose hypotheses is evaluated by gridding the unit spheres where a robust, coarse pose estimate is identified at the mode of the pdf. Further, a \"refining\" pdf of the geodesic distance from the coarse pose estimate is constructed for the hypotheses within a grid containing the coarse estimate. A refined pose estimate is obtained by averaging the low-noise hypotheses in the neighbourhood of the mode of refining pdf. Pose estimation results of the proposed method are compared with RANSAC and nonlinear mean-shift (NMS) algorithms using the Oxford Corridor sequence and the robustness to feature outliers, image noise rejection, and scalability to number of features is analyzed using the synthetic data experiments. Processing time comparison with the RANSAC and NMS algorithms indicate that the deterministic time requirement of the proposed and NMS algorithms is amenable to a variety of visual servo control applications.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122141901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Visual-teach-and-repeat (VT&R) systems have proven extremely useful for practical robot autonomy where the global positioning system is either unavailable or unreliable, examples include tramming for underground mining using a planar laser scanner as well as a return-to-lander function for planetary exploration using a stereo-or laser-based camera. By embedding local appearance/metric information along an arbitrarily long path, it becomes possible to re-drive the path without the need for a single privileged coordinate frame and using only modest computational resources. For a certain class of long-term autonomy problems (e.g., repeatable long-range driving), VT&R appears to offer a simple yet scalable solution. Beyond single paths, we envision that networks of reusable paths could be established and shared from one robot to another to enable practical tasks such as surveillance, delivery (e.g., mail, hospitals, factories, warehouses), worksite operations (e.g., construction, mining), and autonomous roadways. However, for lifelong operations on reusable paths, robustness to a variety of environmental changes, both transient and permanent, is required. In this paper, we relate our experiences and lessons learned with the three above-mentioned implementations of VT&R systems. Based on this, we enumerate both the benefits and challenges of reusable paths that we see moving forwards. We discuss one such challenge, lighting-invariance, in detail and present our progess in overcoming it.
{"title":"Exploiting Reusable Paths in Mobile Robotics: Benefits and Challenges for Long-term Autonomy","authors":"T. Barfoot, B. Stenning, P. Furgale, C. McManus","doi":"10.1109/CRV.2012.58","DOIUrl":"https://doi.org/10.1109/CRV.2012.58","url":null,"abstract":"Visual-teach-and-repeat (VT&R) systems have proven extremely useful for practical robot autonomy where the global positioning system is either unavailable or unreliable, examples include tramming for underground mining using a planar laser scanner as well as a return-to-lander function for planetary exploration using a stereo-or laser-based camera. By embedding local appearance/metric information along an arbitrarily long path, it becomes possible to re-drive the path without the need for a single privileged coordinate frame and using only modest computational resources. For a certain class of long-term autonomy problems (e.g., repeatable long-range driving), VT&R appears to offer a simple yet scalable solution. Beyond single paths, we envision that networks of reusable paths could be established and shared from one robot to another to enable practical tasks such as surveillance, delivery (e.g., mail, hospitals, factories, warehouses), worksite operations (e.g., construction, mining), and autonomous roadways. However, for lifelong operations on reusable paths, robustness to a variety of environmental changes, both transient and permanent, is required. In this paper, we relate our experiences and lessons learned with the three above-mentioned implementations of VT&R systems. Based on this, we enumerate both the benefits and challenges of reusable paths that we see moving forwards. We discuss one such challenge, lighting-invariance, in detail and present our progess in overcoming it.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125191545","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We introduce a new method for distributed object sorting by a swarm of robots. The patch sorting task involves pushing randomly distributed objects into homogeneous clusters. Most existing methods do not make use of vision and are therefore restricted to sensing the objects that lie immediately in front of the robot. We utilize vision both to sense the presence of a cluster and judge its homogeneity, and to seek out distant clusters or isolated objects to pick up. The objects to be sorted are coloured pucks. We present results using a realistic simulation which shows that a simple guidance strategy based on the size of distant clusters can dramatically accelerate the sorting process.
{"title":"Accelerated Patch Sorting by a Robotic Swarm","authors":"A. Vardy","doi":"10.1109/CRV.2012.48","DOIUrl":"https://doi.org/10.1109/CRV.2012.48","url":null,"abstract":"We introduce a new method for distributed object sorting by a swarm of robots. The patch sorting task involves pushing randomly distributed objects into homogeneous clusters. Most existing methods do not make use of vision and are therefore restricted to sensing the objects that lie immediately in front of the robot. We utilize vision both to sense the presence of a cluster and judge its homogeneity, and to seek out distant clusters or isolated objects to pick up. The objects to be sorted are coloured pucks. We present results using a realistic simulation which shows that a simple guidance strategy based on the size of distant clusters can dramatically accelerate the sorting process.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125099605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We present an extension to the Reciprocal Velocity Obstacles (RVO) approach to multi-robot collision avoidance with the aim of alleviating the problem of congestion caused by symmetrical situations in dense conditions. We show that in a resource transportation task RVO robots are unable to make progress due to crowds of robots with opposing navigation goals at source and sink. We introduce Biased Reciprocal Velocity Obstacles (BRVO), which breaks the symmetry among robots by giving priority to the robots leaving a task-related place of interest. BRVO is compared to RVO in two experiments and it is shown that BRVO is able to resolve the congestion much more quickly than RVO.
{"title":"BRaVO: Biased Reciprocal Velocity Obstacles Break Symmetry in Dense Robot Populations","authors":"Seyed Abbas Sadat, R. Vaughan","doi":"10.1109/CRV.2012.65","DOIUrl":"https://doi.org/10.1109/CRV.2012.65","url":null,"abstract":"We present an extension to the Reciprocal Velocity Obstacles (RVO) approach to multi-robot collision avoidance with the aim of alleviating the problem of congestion caused by symmetrical situations in dense conditions. We show that in a resource transportation task RVO robots are unable to make progress due to crowds of robots with opposing navigation goals at source and sink. We introduce Biased Reciprocal Velocity Obstacles (BRVO), which breaks the symmetry among robots by giving priority to the robots leaving a task-related place of interest. BRVO is compared to RVO in two experiments and it is shown that BRVO is able to resolve the congestion much more quickly than RVO.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130101057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
There has been growing interest in the use of binary-valued features, such as BRIEF, ORB, and BRISK for efficient local feature matching. These binary features have several advantages over vector-based features as they can be faster to compute, more compact to store, and more efficient to compare. Although it is fast to compute the Hamming distance between pairs of binary features, particularly on modern architectures, it can still be too slow to use linear search in the case of large datasets. For vector-based features, such as SIFT and SURF, the solution has been to use approximate nearest-neighbor search, but these existing algorithms are not suitable for binary features. In this paper we introduce a new algorithm for approximate matching of binary features, based on priority search of multiple hierarchical clustering trees. We compare this to existing alternatives, and show that it performs well for large datasets, both in terms of speed and memory efficiency.
{"title":"Fast Matching of Binary Features","authors":"Marius Muja, D. Lowe","doi":"10.1109/CRV.2012.60","DOIUrl":"https://doi.org/10.1109/CRV.2012.60","url":null,"abstract":"There has been growing interest in the use of binary-valued features, such as BRIEF, ORB, and BRISK for efficient local feature matching. These binary features have several advantages over vector-based features as they can be faster to compute, more compact to store, and more efficient to compare. Although it is fast to compute the Hamming distance between pairs of binary features, particularly on modern architectures, it can still be too slow to use linear search in the case of large datasets. For vector-based features, such as SIFT and SURF, the solution has been to use approximate nearest-neighbor search, but these existing algorithms are not suitable for binary features. In this paper we introduce a new algorithm for approximate matching of binary features, based on priority search of multiple hierarchical clustering trees. We compare this to existing alternatives, and show that it performs well for large datasets, both in terms of speed and memory efficiency.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129184464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The problem of estimating the pose of metallic objects with shiny surfaces is studied. A new application has been developed using state-of-the-art 3D object segmentation (euclidean clustering) and pose estimation (ICP) methods. We analyze the planar surfaces of the metallic objects in 3D laser scanner data. First we segment these planar objects using euclidean clustering based on surface normals. Thereafter to estimate the pose of these segmented objects we compute Fast Point Feature Histograms (FPFH) descriptors. Finally we use an ICP algorithm that computes the rigid transformation with Singular Value Decomposition(SVD). Two different round of experiments are conducted:-one for the clustering and the other one for the pose estimation. We present the experimental results and analysis along with the possible application scenario and future work.
{"title":"Segmentation and Pose Estimation of Planar Metallic Objects","authors":"Haider Ali, Nadia Figueroa","doi":"10.1109/CRV.2012.56","DOIUrl":"https://doi.org/10.1109/CRV.2012.56","url":null,"abstract":"The problem of estimating the pose of metallic objects with shiny surfaces is studied. A new application has been developed using state-of-the-art 3D object segmentation (euclidean clustering) and pose estimation (ICP) methods. We analyze the planar surfaces of the metallic objects in 3D laser scanner data. First we segment these planar objects using euclidean clustering based on surface normals. Thereafter to estimate the pose of these segmented objects we compute Fast Point Feature Histograms (FPFH) descriptors. Finally we use an ICP algorithm that computes the rigid transformation with Singular Value Decomposition(SVD). Two different round of experiments are conducted:-one for the clustering and the other one for the pose estimation. We present the experimental results and analysis along with the possible application scenario and future work.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129681650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a simple calibration approach to height estimation based on single view metrology. Instead of calibrating intrinsic or extrinsic parameters, our approach aims to estimate the vanishing points of a stationary camera. The calibration process is formulated as an optimization problem with a novel objective function, in which twelve parameters for estimating vanishing points are defined. The genetic algorithm with Cauchy mutation operator is further used for obtaining robust results. The major advantages of the proposed approach are: 1) it is easy to setup since only a cubic box and some optional line segments are required, and 2) it works without camera's intrinsic parameters. Experimental results show the effectiveness of the proposed approach with digital and analog video cameras.
{"title":"A Simple Calibration Approach to Single View Height Estimation","authors":"Kual-Zheng Lee","doi":"10.1109/CRV.2012.29","DOIUrl":"https://doi.org/10.1109/CRV.2012.29","url":null,"abstract":"This paper presents a simple calibration approach to height estimation based on single view metrology. Instead of calibrating intrinsic or extrinsic parameters, our approach aims to estimate the vanishing points of a stationary camera. The calibration process is formulated as an optimization problem with a novel objective function, in which twelve parameters for estimating vanishing points are defined. The genetic algorithm with Cauchy mutation operator is further used for obtaining robust results. The major advantages of the proposed approach are: 1) it is easy to setup since only a cubic box and some optional line segments are required, and 2) it works without camera's intrinsic parameters. Experimental results show the effectiveness of the proposed approach with digital and analog video cameras.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126388567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper adresses the issue of calibrating multiple cameras on a mobile platform. Due to decreasing sensor prices and increasing processing performance, the use of multiple cameras in vehicles becomes an attractive possibility for environment perception. To avoid restrictions relating to the camera arrangement, we focus on non-overlapping camera configurations. Hence, we resign the usage of corresponding features between the cameras. The hand-eye calibration technique based on visual odometry is basically able to solve this problem by exploiting the cameras' motions. However, this technique suffers from inaccuracies in motion estimation. Especially the absolute magnitudes of the translational velocities of each camera are essential for a successful calibration. This contribution presents a novel approach to solve the hand eye calibration problem for two cameras on a mobile platform with non-overlapping fields of view. The so-called motion adjustment simultaneously estimates the extrinsic parameters up to scale as well as the reltive motion magnitudes. Results with simulated and real data are presented.
{"title":"Motion Adjustment for Extrinsic Calibration of Cameras with Non-overlapping Views","authors":"F. Pagel","doi":"10.1109/CRV.2012.20","DOIUrl":"https://doi.org/10.1109/CRV.2012.20","url":null,"abstract":"This paper adresses the issue of calibrating multiple cameras on a mobile platform. Due to decreasing sensor prices and increasing processing performance, the use of multiple cameras in vehicles becomes an attractive possibility for environment perception. To avoid restrictions relating to the camera arrangement, we focus on non-overlapping camera configurations. Hence, we resign the usage of corresponding features between the cameras. The hand-eye calibration technique based on visual odometry is basically able to solve this problem by exploiting the cameras' motions. However, this technique suffers from inaccuracies in motion estimation. Especially the absolute magnitudes of the translational velocities of each camera are essential for a successful calibration. This contribution presents a novel approach to solve the hand eye calibration problem for two cameras on a mobile platform with non-overlapping fields of view. The so-called motion adjustment simultaneously estimates the extrinsic parameters up to scale as well as the reltive motion magnitudes. Results with simulated and real data are presented.","PeriodicalId":372951,"journal":{"name":"2012 Ninth Conference on Computer and Robot Vision","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117194957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：