In this paper we present our approach to 3D surface reconstruction from large sparse range data sets. In space robotics constructing an accurate model of the environment is very important for a variety of reasons. In particular, the constructed model can be used for: safe tele-operation, path planning, planetary exploration and mapping of points of interest. Our approach is based on acquiring range scans from different view-points with overlapping regions, merge them together into a single data set, and fit a triangular mesh on the merged data points. We demonstrate the effectiveness of our approach in a path planning scenario and also by creating the accessibility map for a portion of the Mars Yard located in the Canadian Space Agency.
{"title":"3D reconstruction of environments for planetary exploration","authors":"S. Gemme, J. Bakambu, Ioannis M. Rekleitis","doi":"10.1109/CRV.2005.3","DOIUrl":"https://doi.org/10.1109/CRV.2005.3","url":null,"abstract":"In this paper we present our approach to 3D surface reconstruction from large sparse range data sets. In space robotics constructing an accurate model of the environment is very important for a variety of reasons. In particular, the constructed model can be used for: safe tele-operation, path planning, planetary exploration and mapping of points of interest. Our approach is based on acquiring range scans from different view-points with overlapping regions, merge them together into a single data set, and fit a triangular mesh on the merged data points. We demonstrate the effectiveness of our approach in a path planning scenario and also by creating the accessibility map for a portion of the Mars Yard located in the Canadian Space Agency.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127700157","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}
Aldrin Barreto-Flores, L. A. Robles, Rosa Maria Morales Tepalt, J. Aragon
This paper describes a method for the temporal analysis of texture in colposcopy. The objective is to find temporal texture patterns in order to detect precursory cancer lesions analyzing colposcopy video frames. Preprocessing of the frames is necessary in order to deal with patient movement and non uniform illumination. We use a stabilization algorithm based in a homography and to eliminate incorrect transformations between frames. Illumination correction is done using a local pixel transformation based in the mean around a small window. Temporal reaction after acetic acid application in the cervix is evaluated through the use of a co-occurrence matrix in different regions of the cervix. The reaction is plotted and analyzed through time. Different patterns for normal and abnormal regions are found by this temporal texture analysis showing the possibility to detect important lesions. The proposed method uses standard colposcopy equipment and it was tested using sequences obtained from different patients.
{"title":"Identifying precursory cancer lesions using temporal texture analysis","authors":"Aldrin Barreto-Flores, L. A. Robles, Rosa Maria Morales Tepalt, J. Aragon","doi":"10.1109/CRV.2005.48","DOIUrl":"https://doi.org/10.1109/CRV.2005.48","url":null,"abstract":"This paper describes a method for the temporal analysis of texture in colposcopy. The objective is to find temporal texture patterns in order to detect precursory cancer lesions analyzing colposcopy video frames. Preprocessing of the frames is necessary in order to deal with patient movement and non uniform illumination. We use a stabilization algorithm based in a homography and to eliminate incorrect transformations between frames. Illumination correction is done using a local pixel transformation based in the mean around a small window. Temporal reaction after acetic acid application in the cervix is evaluated through the use of a co-occurrence matrix in different regions of the cervix. The reaction is plotted and analyzed through time. Different patterns for normal and abnormal regions are found by this temporal texture analysis showing the possibility to detect important lesions. The proposed method uses standard colposcopy equipment and it was tested using sequences obtained from different patients.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123217028","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}
Spacecraft docking using vision is a challenging task. Not least among the problems encountered is the need to visually localize the docking target. Here we consider the task of adapting the local illumination to assist in this docking. An online approach is developed that combines images obtained under different exposure and lighting conditions into a single image upon which docking decisions can be made. This method is designed to be used within an intelligent controller that automatically adjusts lighting and image acquisition in order to obtain the "best" possible composite view of the target for further image processing.
{"title":"Entropy-based image merging","authors":"A. German, M. Jenkin, Y. Lespérance","doi":"10.1109/CRV.2005.38","DOIUrl":"https://doi.org/10.1109/CRV.2005.38","url":null,"abstract":"Spacecraft docking using vision is a challenging task. Not least among the problems encountered is the need to visually localize the docking target. Here we consider the task of adapting the local illumination to assist in this docking. An online approach is developed that combines images obtained under different exposure and lighting conditions into a single image upon which docking decisions can be made. This method is designed to be used within an intelligent controller that automatically adjusts lighting and image acquisition in order to obtain the \"best\" possible composite view of the target for further image processing.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123227887","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}
Degradation of images of outdoor scenes caused by varying conditions of visibility can be exploited in order to get information on the scene. We propose two new methods for detecting occlusion edges between textured areas of a scene. These methods are based on the use of partial derivatives of two images acquired under different conditions of visibility. They were validated on images of both synthetic and real scenes.
{"title":"Detection of occlusion edges from the derivatives of weather degraded images","authors":"Daniel Lévesque, F. Deschênes","doi":"10.1109/CRV.2005.35","DOIUrl":"https://doi.org/10.1109/CRV.2005.35","url":null,"abstract":"Degradation of images of outdoor scenes caused by varying conditions of visibility can be exploited in order to get information on the scene. We propose two new methods for detecting occlusion edges between textured areas of a scene. These methods are based on the use of partial derivatives of two images acquired under different conditions of visibility. They were validated on images of both synthetic and real scenes.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121754010","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 and new approach to robot exploration and mapping using a team of cooperative robots. This approach aims to exploit the increase in sensor data that multiple robots offer to improve efficiency, accuracy and detail in maps created, and also the lower cost in employing a group of inexpensive robots. The exploration technique involves covering an area as efficiently as possible while cooperating to estimate each other's positions and orientations. The ability to observe objects of interest from a number of viewpoints and combine this data means that cooperative robots can localize objects and estimate their shape in cluttered real world scenes. Robots in the system act as social agents, and are motivated to cooperate by a desire to increase their own utility. Within this society, robots form coalitions to complete tasks that arise which require input from multiple robots. The coalitions involve the adoption of certain roles or behaviors on the part of the different robots to carry out these tasks.
{"title":"Collaborative exploration for a group of self-interested robots","authors":"M. Schukat, Declan O'Beirne","doi":"10.1109/CRV.2005.25","DOIUrl":"https://doi.org/10.1109/CRV.2005.25","url":null,"abstract":"This paper presents and new approach to robot exploration and mapping using a team of cooperative robots. This approach aims to exploit the increase in sensor data that multiple robots offer to improve efficiency, accuracy and detail in maps created, and also the lower cost in employing a group of inexpensive robots. The exploration technique involves covering an area as efficiently as possible while cooperating to estimate each other's positions and orientations. The ability to observe objects of interest from a number of viewpoints and combine this data means that cooperative robots can localize objects and estimate their shape in cluttered real world scenes. Robots in the system act as social agents, and are motivated to cooperate by a desire to increase their own utility. Within this society, robots form coalitions to complete tasks that arise which require input from multiple robots. The coalitions involve the adoption of certain roles or behaviors on the part of the different robots to carry out these tasks.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129580150","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}
A fundamental task in computer vision is that of determining the position and orientation of a moving camera relative to an observed object or scene. Many such visual tracking algorithms have been proposed in the computer vision, artificial intelligence and robotics literature over the past 30 years. Predominantly, these remain un-validated since the ground-truth camera positions and orientations at each frame in a video sequence are not available for comparison with the outputs of the proposed vision systems. A method is presented for generating real visual test data with complete underlying ground-truth. The method enables the production of long video sequences, filmed along complicated six degree of freedom trajectories, featuring a variety of objects, in a variety of different visibility conditions, for which complete ground-truth data is known including the camera position and orientation at every image frame, intrinsic camera calibration data, a lens distortion model and models of the viewed objects.
{"title":"Video with ground-truth for validation of visual registration, tracking and navigation algorithms","authors":"R. Stolkin, A. Greig, J. Gilby","doi":"10.1109/CRV.2005.86","DOIUrl":"https://doi.org/10.1109/CRV.2005.86","url":null,"abstract":"A fundamental task in computer vision is that of determining the position and orientation of a moving camera relative to an observed object or scene. Many such visual tracking algorithms have been proposed in the computer vision, artificial intelligence and robotics literature over the past 30 years. Predominantly, these remain un-validated since the ground-truth camera positions and orientations at each frame in a video sequence are not available for comparison with the outputs of the proposed vision systems. A method is presented for generating real visual test data with complete underlying ground-truth. The method enables the production of long video sequences, filmed along complicated six degree of freedom trajectories, featuring a variety of objects, in a variety of different visibility conditions, for which complete ground-truth data is known including the camera position and orientation at every image frame, intrinsic camera calibration data, a lens distortion model and models of the viewed objects.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127117830","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}
In this paper, we propose an algorithm that detects and locates natural objects in an outdoor environment using local descriptors. Interest points inside images are detected with a difference of Gaussian (DoG) filter and are then represented using scale invariant local descriptors. Our algorithm learns objects in a weakly supervised manner by clustering similar descriptors together and using those clusters as object classifiers. The intent is to identify stable objects to be used as landmarks for simultaneous localization and mapping (SLAM) of robots. The robot milieu is first identified using a fast environment recognition algorithm and then landmarks are suggested for SLAM that are appropriate for that environment. In our experiments we test our theory on the detection of trees that belong to the plantae pinophyta (pine family). Initial results show that out of 200 test images, our classification yields 85 correct positives, 15 false negatives, 73 correct negatives and 27 false positives.
{"title":"Seeing the trees before the forest [natural object detection]","authors":"Daniel C. Asmar, J. Zelek, Samer M. Abdallah","doi":"10.1109/CRV.2005.71","DOIUrl":"https://doi.org/10.1109/CRV.2005.71","url":null,"abstract":"In this paper, we propose an algorithm that detects and locates natural objects in an outdoor environment using local descriptors. Interest points inside images are detected with a difference of Gaussian (DoG) filter and are then represented using scale invariant local descriptors. Our algorithm learns objects in a weakly supervised manner by clustering similar descriptors together and using those clusters as object classifiers. The intent is to identify stable objects to be used as landmarks for simultaneous localization and mapping (SLAM) of robots. The robot milieu is first identified using a fast environment recognition algorithm and then landmarks are suggested for SLAM that are appropriate for that environment. In our experiments we test our theory on the detection of trees that belong to the plantae pinophyta (pine family). Initial results show that out of 200 test images, our classification yields 85 correct positives, 15 false negatives, 73 correct negatives and 27 false positives.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128406049","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}
In this paper we present a 3D tracking system that is able to identify several characteristic points on the surface of a moving scene. The system is considered as a preliminary step in the interaction of industrial robots with dynamic scenes. Through a new color structured light technique based on a disordered codeword pattern, 3D coordinates of the object surface are extracted and processed. This method recovers 3D information in such a way that the correspondence problem is easily and robustly solved. After establishing a set of feature points, an inter frame window search algorithm is carried out to solve the tracking problem. A controlled experimental setup has been built in our lab composed by a 2 DOF mobile platform, a light structured sensor and a manipulator robot. The experimentation has been performed on medium spatial resolution and for soft movement specifications giving promising results.
{"title":"3D feature tracking using a dynamic structured light system","authors":"A. Adán, F. Molina, A. Vázquez, Luis Morena","doi":"10.1109/CRV.2005.1","DOIUrl":"https://doi.org/10.1109/CRV.2005.1","url":null,"abstract":"In this paper we present a 3D tracking system that is able to identify several characteristic points on the surface of a moving scene. The system is considered as a preliminary step in the interaction of industrial robots with dynamic scenes. Through a new color structured light technique based on a disordered codeword pattern, 3D coordinates of the object surface are extracted and processed. This method recovers 3D information in such a way that the correspondence problem is easily and robustly solved. After establishing a set of feature points, an inter frame window search algorithm is carried out to solve the tracking problem. A controlled experimental setup has been built in our lab composed by a 2 DOF mobile platform, a light structured sensor and a manipulator robot. The experimentation has been performed on medium spatial resolution and for soft movement specifications giving promising results.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128838181","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}
In this paper, we extend the new photometric stereo method of Hertzmenn and Seitz that uses many images of an object together with a calibration object. For each point in the registered collection of images, we have a large number of brightness values. Photometric stereo finds a similar collection of brightness values from the calibration object and overdetermines the surface normal. With a large number of images, finding similar brightnesses becomes costly search in high dimensions. To speed up the search, we apply locality sensitive high dimensional hashing (LSH) to compute the irregular target object's surface orientation. The experimental results of a simplified photometric stereo experiment show consistent results in surface orientation. LSH can be implemented very efficiently and offers the possibility of practical photometric stereo computation with a large number of images.
{"title":"Photometric stereo via locality sensitive high-dimension hashing","authors":"Lin Zhong, J. Little","doi":"10.1109/CRV.2005.61","DOIUrl":"https://doi.org/10.1109/CRV.2005.61","url":null,"abstract":"In this paper, we extend the new photometric stereo method of Hertzmenn and Seitz that uses many images of an object together with a calibration object. For each point in the registered collection of images, we have a large number of brightness values. Photometric stereo finds a similar collection of brightness values from the calibration object and overdetermines the surface normal. With a large number of images, finding similar brightnesses becomes costly search in high dimensions. To speed up the search, we apply locality sensitive high dimensional hashing (LSH) to compute the irregular target object's surface orientation. The experimental results of a simplified photometric stereo experiment show consistent results in surface orientation. LSH can be implemented very efficiently and offers the possibility of practical photometric stereo computation with a large number of images.","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115507909","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 number of new views and techniques claimed to be very important for the problem of face recognition in video (FRiV). First, a clear differentiation is made between photographic facial data and video-acquired facial data as being two different modalities: one providing hard biometrics, the other providing softer biometrics. Second, faces which have the resolution of at least 12 pixels between the eyes are shown to be recognizable by computers just as they are by humans. As a way to deal with low resolution and quality of each individual video frame, the paper offers to use the neuro-associative principle employed by human brain, according to which both memorization and recognition of data are done based on a flow of frames rather than on one frame: synaptic plasticity provides a way to memorize from a sequence, while the collective decision making over time is very suitable for recognition of a sequence. As a benchmark for FRiV approaches, the paper introduces the IIT-NRC video-based database of faces which consists of pairs of low-resolution video clips of unconstrained facial motions. The recognition rate of over 95%, which we achieve on this database, as well as the results obtained on real-time annotation of people on TV allow us to believe that the proposed framework brings us closer to the ultimate benchmark for the FRiV approaches, which is "if you are able to recognize a person, so should the computer".
{"title":"Video-based framework for face recognition in video","authors":"D. Gorodnichy","doi":"10.1109/CRV.2005.87","DOIUrl":"https://doi.org/10.1109/CRV.2005.87","url":null,"abstract":"This paper presents a number of new views and techniques claimed to be very important for the problem of face recognition in video (FRiV). First, a clear differentiation is made between photographic facial data and video-acquired facial data as being two different modalities: one providing hard biometrics, the other providing softer biometrics. Second, faces which have the resolution of at least 12 pixels between the eyes are shown to be recognizable by computers just as they are by humans. As a way to deal with low resolution and quality of each individual video frame, the paper offers to use the neuro-associative principle employed by human brain, according to which both memorization and recognition of data are done based on a flow of frames rather than on one frame: synaptic plasticity provides a way to memorize from a sequence, while the collective decision making over time is very suitable for recognition of a sequence. As a benchmark for FRiV approaches, the paper introduces the IIT-NRC video-based database of faces which consists of pairs of low-resolution video clips of unconstrained facial motions. The recognition rate of over 95%, which we achieve on this database, as well as the results obtained on real-time annotation of people on TV allow us to believe that the proposed framework brings us closer to the ultimate benchmark for the FRiV approaches, which is \"if you are able to recognize a person, so should the computer\".","PeriodicalId":307318,"journal":{"name":"The 2nd Canadian Conference on Computer and Robot Vision (CRV'05)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115730460","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：