Han Hu, Liupeng Su, Shunfu Mao, Min Chen, Guoqiang Pan, Bo Xu, Qing Zhu
Deep‐learning methods have demonstrated promising performance in multi‐view stereo (MVS) applications. However, it remains challenging to apply a geometrical prior on the adaptive matching windows to achieve efficient three‐dimensional reconstruction. To address this problem, this paper proposes a learnable adaptive region aggregation method based on deformable convolutional networks (DCNs), which is integrated into the feature extraction workflow for MVSNet method that uses coarse‐to‐fine structure. Following the conventional pipeline of MVSNet, a DCN is used to densely estimate and apply transformations in our feature extractor, which is composed of a deformable feature pyramid network (DFPN). Furthermore, we introduce a dedicated offset regulariser to promote the convergence of the learnable offsets of the DCN. The effectiveness of the proposed DFPN is validated through quantitative and qualitative evaluations on the BlendedMVS and Tanks and Temples benchmark datasets within a cross‐dataset evaluation setting.
深度学习方法在多视点立体(MVS)应用中表现出了良好的性能。然而，如何在自适应匹配窗口上应用几何先验来实现有效的三维重建仍然是一个挑战。为了解决这一问题，本文提出了一种基于可变形卷积网络(DCNs)的可学习自适应区域聚合方法，并将其集成到使用粗-细结构的MVSNet方法的特征提取工作流程中。在MVSNet传统管道的基础上，利用DCN对可变形特征金字塔网络(DFPN)构成的特征提取器进行密集估计和变换。此外，我们还引入了一个专用的偏移校正器来促进DCN的可学习偏移的收敛性。通过在交叉数据集评估设置中对BlendedMVS和Tanks and Temples基准数据集进行定量和定性评估，验证了所提出的DFPN的有效性。
{"title":"Adaptive region aggregation for multi‐view stereo matching using deformable convolutional networks","authors":"Han Hu, Liupeng Su, Shunfu Mao, Min Chen, Guoqiang Pan, Bo Xu, Qing Zhu","doi":"10.1111/phor.12459","DOIUrl":"https://doi.org/10.1111/phor.12459","url":null,"abstract":"Deep‐learning methods have demonstrated promising performance in multi‐view stereo (MVS) applications. However, it remains challenging to apply a geometrical prior on the adaptive matching windows to achieve efficient three‐dimensional reconstruction. To address this problem, this paper proposes a learnable adaptive region aggregation method based on deformable convolutional networks (DCNs), which is integrated into the feature extraction workflow for MVSNet method that uses coarse‐to‐fine structure. Following the conventional pipeline of MVSNet, a DCN is used to densely estimate and apply transformations in our feature extractor, which is composed of a deformable feature pyramid network (DFPN). Furthermore, we introduce a dedicated offset regulariser to promote the convergence of the learnable offsets of the DCN. The effectiveness of the proposed DFPN is validated through quantitative and qualitative evaluations on the BlendedMVS and Tanks and Temples benchmark datasets within a cross‐dataset evaluation setting.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"16 1","pages":"430 - 449"},"PeriodicalIF":0.0,"publicationDate":"2023-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81631377","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}
Elisavet (Ellie) Konstantina Stathopoulou, F. Remondino
3D reconstruction of scenes using multiple images, relying on robust correspondence search and depth estimation, has been thoroughly studied for the two‐view and multi‐view scenarios in recent years. Multi‐view stereo (MVS) algorithms aim to generate a rich, dense 3D model of the scene in the form of a dense point cloud or a triangulated mesh. In a typical MVS pipeline, the robust estimations for the camera poses along with the sparse points obtained from structure from motion (SfM) are used as input. During this process, the depth of generally every pixel of the scene is to be calculated. Several methods, either conventional or, more recently, learning‐based have been developed for solving the correspondence search problem. A vast amount of research exists in the literature using local, global or semi‐global stereomatching approaches, with the PatchMatch algorithm being among the most popular and efficient conventional ones in the last decade. Yet, and despite the widespread evolution of the algorithms, yielding complete, accurate and aesthetically pleasing 3D representations of a scene remains an open issue in real‐world and large‐scale photogrammetric applications. This work aims to provide a concrete survey on the most widely used MVS methods, investigating underlying concepts and challenges. To this end, the theoretical background and relative literature are discussed for both conventional and learning‐based approaches, with a particular focus on close‐range 3D reconstruction applications.
{"title":"A survey on conventional and learning‐based methods for multi‐view stereo","authors":"Elisavet (Ellie) Konstantina Stathopoulou, F. Remondino","doi":"10.1111/phor.12456","DOIUrl":"https://doi.org/10.1111/phor.12456","url":null,"abstract":"3D reconstruction of scenes using multiple images, relying on robust correspondence search and depth estimation, has been thoroughly studied for the two‐view and multi‐view scenarios in recent years. Multi‐view stereo (MVS) algorithms aim to generate a rich, dense 3D model of the scene in the form of a dense point cloud or a triangulated mesh. In a typical MVS pipeline, the robust estimations for the camera poses along with the sparse points obtained from structure from motion (SfM) are used as input. During this process, the depth of generally every pixel of the scene is to be calculated. Several methods, either conventional or, more recently, learning‐based have been developed for solving the correspondence search problem. A vast amount of research exists in the literature using local, global or semi‐global stereomatching approaches, with the PatchMatch algorithm being among the most popular and efficient conventional ones in the last decade. Yet, and despite the widespread evolution of the algorithms, yielding complete, accurate and aesthetically pleasing 3D representations of a scene remains an open issue in real‐world and large‐scale photogrammetric applications. This work aims to provide a concrete survey on the most widely used MVS methods, investigating underlying concepts and challenges. To this end, the theoretical background and relative literature are discussed for both conventional and learning‐based approaches, with a particular focus on close‐range 3D reconstruction applications.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"96 1","pages":"374 - 407"},"PeriodicalIF":0.0,"publicationDate":"2023-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83363640","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 propose an in situ self‐calibration method by detecting and matching intensity features on the local planes in overlapping point clouds based on the Förstner operator. We successfully matched the intensity features from scans at different locations by feature matching on common local planes rather than on the rasterised grids of the horizontal and vertical angles adopted by the affirmed keypoint‐based algorithm. The capability of extracting features from different stations offers the possibility of comprehensive scanner calibration, solving the disadvantage that the existing keypoint‐based methods can only estimate the two‐face‐sensitive model parameters. The proposed algorithm has been tested with a high‐precision panoramic scanner, Leica RTC360, using datasets from a calibration hall and a general working scenario. It has been shown that the proposed approach consistently calibrates the two‐face‐sensitive model parameters with the affirmed keypoint‐based one. For the case of comprehensive calibration with the offset estimated and some angular parameters separated where the previous keypoint‐based one failed, the proposed algorithm achieves an accuracy of 0.16 mm, 2.7″ and 2.1″ in range, azimuth and elevation for the estimated target centres. The proposed algorithm can accurately calibrate two‐face‐sensitive and more comprehensive model parameters without any preparation on‐site, for example, mounting targets.
本文提出了一种基于Förstner算子的重叠点云局部平面强度特征检测与匹配的原位自校准方法。我们通过在共同局部平面上的特征匹配来成功匹配不同位置扫描的强度特征，而不是采用基于确认关键点的算法所采用的水平和垂直角度的光栅网格。从不同台站提取特征的能力为全面的扫描仪校准提供了可能，解决了现有基于关键点的方法只能估计双面敏感模型参数的缺点。该算法已在高精度全景扫描仪徕卡RTC360上进行了测试，使用来自校准大厅的数据集和一般工作场景。结果表明，该方法与确定的基于关键点的模型参数一致地校准了两个面敏感模型参数。对于先前基于关键点的校准失败的偏移量估计和一些角度参数分离的综合校准情况，该算法对估计的目标中心的距离，方位角和仰角精度分别达到0.16 mm, 2.7″和2.1″。该算法可以准确地校准两面敏感的更全面的模型参数，而无需进行任何现场准备，例如安装目标。
{"title":"Automatic calibration of terrestrial laser scanners using intensity features","authors":"Jing Qiao, Tomislav Medic, Andreas Baumann-Ouyang","doi":"10.1111/phor.12454","DOIUrl":"https://doi.org/10.1111/phor.12454","url":null,"abstract":"We propose an in situ self‐calibration method by detecting and matching intensity features on the local planes in overlapping point clouds based on the Förstner operator. We successfully matched the intensity features from scans at different locations by feature matching on common local planes rather than on the rasterised grids of the horizontal and vertical angles adopted by the affirmed keypoint‐based algorithm. The capability of extracting features from different stations offers the possibility of comprehensive scanner calibration, solving the disadvantage that the existing keypoint‐based methods can only estimate the two‐face‐sensitive model parameters. The proposed algorithm has been tested with a high‐precision panoramic scanner, Leica RTC360, using datasets from a calibration hall and a general working scenario. It has been shown that the proposed approach consistently calibrates the two‐face‐sensitive model parameters with the affirmed keypoint‐based one. For the case of comprehensive calibration with the offset estimated and some angular parameters separated where the previous keypoint‐based one failed, the proposed algorithm achieves an accuracy of 0.16 mm, 2.7″ and 2.1″ in range, azimuth and elevation for the estimated target centres. The proposed algorithm can accurately calibrate two‐face‐sensitive and more comprehensive model parameters without any preparation on‐site, for example, mounting targets.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"32 5","pages":"320 - 338"},"PeriodicalIF":0.0,"publicationDate":"2023-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91489037","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 distributed fisheye video surveillance system (DFVSS) can monitor a wide area without blind spots, but it is often affected by the viewpoint discontinuity and space inconsistency of multiple videos in the area. This paper proposes a novel real‐time fisheye video mosaic algorithm for wide‐area surveillance. First, by extending the line photogrammetry theory under central projection to spherical projection, a fisheye video geo‐registration model is established and estimated using orthogonal parallel lines on the ground, so that all videos of DFVSS are in the unified reference system to eliminate the space inconsistency between them. Second, by combining the photogrammetry orthorectification technique with thin‐plate spline transformation, a fisheye video rectification model is established to eliminate serious distortion in geo‐registered fisheye videos and align them accurately. Third, the viewport‐dependent video selection strategy and video look‐up table computation technique are adopted to create a high‐resolution panorama from input fisheye videos in real time. A parking lot of about 0.4 km2 monitored by eight fisheye cameras was selected as the test area. The experimental result shows the line re‐projection error in fisheye videos is about 0.5 pixels, and the overall efficiency, including panorama creation and mapping to the ground as texture, is not <30 fps. It indicates that the proposed algorithm can achieve a good balance between the limitation of video transmission bandwidth and the smooth observation requirement of computer equipment for the panorama, which is of great value for the construction and application of DFVSS.
{"title":"Real‐time mosaic of multiple fisheye surveillance videos based on geo‐registration and rectification","authors":"Jiongli Gao, Jun Wu, Mingyi Huang, Gang Xu","doi":"10.1111/phor.12455","DOIUrl":"https://doi.org/10.1111/phor.12455","url":null,"abstract":"A distributed fisheye video surveillance system (DFVSS) can monitor a wide area without blind spots, but it is often affected by the viewpoint discontinuity and space inconsistency of multiple videos in the area. This paper proposes a novel real‐time fisheye video mosaic algorithm for wide‐area surveillance. First, by extending the line photogrammetry theory under central projection to spherical projection, a fisheye video geo‐registration model is established and estimated using orthogonal parallel lines on the ground, so that all videos of DFVSS are in the unified reference system to eliminate the space inconsistency between them. Second, by combining the photogrammetry orthorectification technique with thin‐plate spline transformation, a fisheye video rectification model is established to eliminate serious distortion in geo‐registered fisheye videos and align them accurately. Third, the viewport‐dependent video selection strategy and video look‐up table computation technique are adopted to create a high‐resolution panorama from input fisheye videos in real time. A parking lot of about 0.4 km2 monitored by eight fisheye cameras was selected as the test area. The experimental result shows the line re‐projection error in fisheye videos is about 0.5 pixels, and the overall efficiency, including panorama creation and mapping to the ground as texture, is not <30 fps. It indicates that the proposed algorithm can achieve a good balance between the limitation of video transmission bandwidth and the smooth observation requirement of computer equipment for the panorama, which is of great value for the construction and application of DFVSS.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"1 1","pages":"339 - 373"},"PeriodicalIF":0.0,"publicationDate":"2023-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85449142","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}
Haonan Dong, Jian Yao, Ye Gong, Li Li, Shaosheng Cao, Yuxuan Li
Fisheye camera calibration is an essential task in photogrammetry. However, previous calibration patterns and the robustness of the adjoint processing methods are limited due to the fisheye distortion and various lighting. This problem leads to additional manual intervention in the data collection. Moreover, it is arduous to accurately detect the board target under fisheye's distortion. To increase the robustness in this task, we present a novel encoded board “Meta‐Board” and a learning‐based target detection method. Additionally, an automatic image orthorectification is integrated to alleviate the distortion effect on the target iteratively until convergence. A low‐cost control field with the proposed boards is built for the experiment. Results on both virtual and real camera lenses and multi‐camera rigs show that our method can be robustly used in calibrating the fisheye camera and reaches state‐of‐the‐art accuracy.
{"title":"Learning‐based encoded target detection on iteratively orthorectified images for accurate fisheye calibration","authors":"Haonan Dong, Jian Yao, Ye Gong, Li Li, Shaosheng Cao, Yuxuan Li","doi":"10.1111/phor.12453","DOIUrl":"https://doi.org/10.1111/phor.12453","url":null,"abstract":"Fisheye camera calibration is an essential task in photogrammetry. However, previous calibration patterns and the robustness of the adjoint processing methods are limited due to the fisheye distortion and various lighting. This problem leads to additional manual intervention in the data collection. Moreover, it is arduous to accurately detect the board target under fisheye's distortion. To increase the robustness in this task, we present a novel encoded board “Meta‐Board” and a learning‐based target detection method. Additionally, an automatic image orthorectification is integrated to alleviate the distortion effect on the target iteratively until convergence. A low‐cost control field with the proposed boards is built for the experiment. Results on both virtual and real camera lenses and multi‐camera rigs show that our method can be robustly used in calibrating the fisheye camera and reaches state‐of‐the‐art accuracy.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"156 1","pages":"297 - 319"},"PeriodicalIF":0.0,"publicationDate":"2023-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77381646","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}
Leaf area index (LAI) is one of the key parameters of vegetation structure, which can be applied in monitoring vegetation growth status. Currently, abundant spatial information (e.g., textural information), provided by the developing remote sensing satellite techniques, could boost the accuracy of LAI estimation. Thus, the performance of spectral and textural information must be evaluated for different vegetation types of LAI estimation in different surface types. In this study, different spectral vegetation indices (SVIs) and grey‐level co‐occurrence matrix‐based textural variables under different moving window sizes were extracted from Landsat TM satellite data. First, the ability of different types of SVIs for LAI estimation in different surface types was analysed. Subsequently, the effect of different texture variables with different moving window sizes towards LAI estimation accuracy in different vegetation types was explored. Lastly, the performance of SVIs combined with textural information for the LAI estimation in different vegetation types was evaluated. Results indicated that SVIs performed better for LAI estimation in the homogeneous region than that in the heterogeneous region, and difference vegetation index was more remarkable for LAI estimation in different vegetation types than other SVIs. In addition, variations in texture variables and moving window sizes had a large influence on LAI estimation of natural vegetation with high canopy heterogeneity. SVI combined with textural information can efficiently improve the accuracy of LAI estimation in different vegetation types (R2 = 0.672, 0.455 and 0.523 for meadow, shrub and cantaloupe, respectively.) compared with SVI alone (R2 = 0.189, 0.064 and 0.431 for meadow, shrub and cantaloupe, respectively.). Especially for natural vegetation (meadow, shrub), the addition of textural information can greatly improve the accuracy of LAI estimation.
{"title":"Exploring the performance of spectral and textural information for leaf area index estimation with homogeneous and heterogeneous surfaces","authors":"Yangyang Zhang, Xu Han, Jian Yang","doi":"10.1111/phor.12450","DOIUrl":"https://doi.org/10.1111/phor.12450","url":null,"abstract":"Leaf area index (LAI) is one of the key parameters of vegetation structure, which can be applied in monitoring vegetation growth status. Currently, abundant spatial information (e.g., textural information), provided by the developing remote sensing satellite techniques, could boost the accuracy of LAI estimation. Thus, the performance of spectral and textural information must be evaluated for different vegetation types of LAI estimation in different surface types. In this study, different spectral vegetation indices (SVIs) and grey‐level co‐occurrence matrix‐based textural variables under different moving window sizes were extracted from Landsat TM satellite data. First, the ability of different types of SVIs for LAI estimation in different surface types was analysed. Subsequently, the effect of different texture variables with different moving window sizes towards LAI estimation accuracy in different vegetation types was explored. Lastly, the performance of SVIs combined with textural information for the LAI estimation in different vegetation types was evaluated. Results indicated that SVIs performed better for LAI estimation in the homogeneous region than that in the heterogeneous region, and difference vegetation index was more remarkable for LAI estimation in different vegetation types than other SVIs. In addition, variations in texture variables and moving window sizes had a large influence on LAI estimation of natural vegetation with high canopy heterogeneity. SVI combined with textural information can efficiently improve the accuracy of LAI estimation in different vegetation types (R2 = 0.672, 0.455 and 0.523 for meadow, shrub and cantaloupe, respectively.) compared with SVI alone (R2 = 0.189, 0.064 and 0.431 for meadow, shrub and cantaloupe, respectively.). Especially for natural vegetation (meadow, shrub), the addition of textural information can greatly improve the accuracy of LAI estimation.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"1 1","pages":"233 - 251"},"PeriodicalIF":0.0,"publicationDate":"2023-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83710884","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}
Ye Gong, Pengwei Zhou, Yu‐ye Liu, Haonan Dong, Li Li, Jian Yao
Structure from motion (SfM) is used to recover camera poses and the sparse structure of real scenes from multiview images. SfM methods construct a view‐graph from the matching relationships of images. Redundancy and incorrect edges are usually observed in it. Redundancy inhibits the efficiency and incorrect edges result in the misalignment of structures. In addition, the uneven distribution of vertices usually affects the global accuracy. To address these problems, we propose a coarse‐to‐fine approach in which the poses of an extracted key‐subset of images are first computed and then all remaining images are oriented. The core of this approach is view‐graph key‐subset extraction, which not only prunes redundant data and incorrect edges but also obtains properly distributed key‐subset vertices. The extraction approach is based on a replaceability score and an iteration‐update strategy. In this way, only vertices with high SfM importance are preserved in the key‐subset. Different public datasets are used to evaluate our approach. Due to the absence of ground‐truth camera poses in large‐scale datasets, we present new datasets with accurate camera poses and point clouds. The results demonstrate that our approach greatly increases the efficiency of SfM. Furthermore, the robustness and accuracy can be improved.
{"title":"View‐graph key‐subset extraction for efficient and robust structure from motion","authors":"Ye Gong, Pengwei Zhou, Yu‐ye Liu, Haonan Dong, Li Li, Jian Yao","doi":"10.1111/phor.12451","DOIUrl":"https://doi.org/10.1111/phor.12451","url":null,"abstract":"Structure from motion (SfM) is used to recover camera poses and the sparse structure of real scenes from multiview images. SfM methods construct a view‐graph from the matching relationships of images. Redundancy and incorrect edges are usually observed in it. Redundancy inhibits the efficiency and incorrect edges result in the misalignment of structures. In addition, the uneven distribution of vertices usually affects the global accuracy. To address these problems, we propose a coarse‐to‐fine approach in which the poses of an extracted key‐subset of images are first computed and then all remaining images are oriented. The core of this approach is view‐graph key‐subset extraction, which not only prunes redundant data and incorrect edges but also obtains properly distributed key‐subset vertices. The extraction approach is based on a replaceability score and an iteration‐update strategy. In this way, only vertices with high SfM importance are preserved in the key‐subset. Different public datasets are used to evaluate our approach. Due to the absence of ground‐truth camera poses in large‐scale datasets, we present new datasets with accurate camera poses and point clouds. The results demonstrate that our approach greatly increases the efficiency of SfM. Furthermore, the robustness and accuracy can be improved.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"41 1","pages":"252 - 296"},"PeriodicalIF":0.0,"publicationDate":"2023-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87132135","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 wide variety of sensors and systems available on the market for collecting spatial data makes the evaluation of provided information, calibration of sensors and benchmarking of systems a critical task. It is also an important scientific issue for many professionals. In daily work, the assessment of algorithms and sensors for collecting and generating spatial data resources is a crucial issue for academic institutions, research centres, national mapping and cadastral agencies, and all professionals handling geospatial data. The GEOBENCH workshop is therefore appropriate for those willing to extend their knowledge in the fields of photogrammetry and remote sensing – present evaluations of algorithms and sensors in the sector as well as new benchmarks. The workshop is a followup of the first successful event held in Warsaw, Poland, in 2019 and will be held in the AGH University of Science and Technology in Krakow, Poland, on 23– 24 October 2023.
{"title":"2023 Asian Conference on Remote Sensing (ACRS)","authors":"","doi":"10.1111/phor.10_12449","DOIUrl":"https://doi.org/10.1111/phor.10_12449","url":null,"abstract":"The wide variety of sensors and systems available on the market for collecting spatial data makes the evaluation of provided information, calibration of sensors and benchmarking of systems a critical task. It is also an important scientific issue for many professionals. In daily work, the assessment of algorithms and sensors for collecting and generating spatial data resources is a crucial issue for academic institutions, research centres, national mapping and cadastral agencies, and all professionals handling geospatial data. The GEOBENCH workshop is therefore appropriate for those willing to extend their knowledge in the fields of photogrammetry and remote sensing – present evaluations of algorithms and sensors in the sector as well as new benchmarks. The workshop is a followup of the first successful event held in Warsaw, Poland, in 2019 and will be held in the AGH University of Science and Technology in Krakow, Poland, on 23– 24 October 2023.","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83324682","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}
{"title":"Academic Track of Foss4g (Free and Open Source Software for Geospatial) 2023","authors":"","doi":"10.1111/phor.3_12449","DOIUrl":"https://doi.org/10.1111/phor.3_12449","url":null,"abstract":"","PeriodicalId":22881,"journal":{"name":"The Photogrammetric Record","volume":"109 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88724565","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：