[{"data":1,"prerenderedAt":756},["ShallowReactive",2],{"method-zhou2016fgr":3},{"method":4,"reference":55,"equipment":76,"figures":83,"results":84},{"id":5,"label":6,"shortName":7,"title":8,"year":9,"era":10,"cluster":11,"scope":12,"keyIdeaZh":13,"keyIdeaEn":14,"fulltextStatus":15,"publicationStatus":16,"recommendation":17,"constructionRelevance":18,"validationEnvironment":19,"strengths":22,"limitations":29,"sensors":34,"platform":38,"estimator":40,"association":41,"timeModel":42,"deskew":42,"loopClosure":43,"globalOptimization":44,"mapRepresentation":45,"prior":46,"outputGeometry":47,"compute":48,"codeUrl":49,"codeLicense":50,"relatedVersions":51},"zhou2016fgr","Zhou et al., 2016","FGR","Fast Global Registration",2016,"classic","C02","registration_component","FGR 先以 FPFH 特徵的雙向最近鄰建立候選對應，再以互為最近鄰檢驗與三元組邊長比例檢驗（τ = 0.9）提高內點比例；之後對這組固定不變的對應直接最佳化單一穩健目標，同時對齊表面並使錯誤對應失效，內迴圈不更新對應，也不做最近點查詢。目標採縮放 Geman-McClure 穩健估計函數，藉 Black-Rangarajan 對偶交替更新線過程變數與位姿，並以逐步非凸化將 μ 從最大表面直徑的平方每四次迭代減半，直到真對應距離門檻的平方。作者報告在合成資料、UWA 與 Choi 等人的場景基準上，精度可比或優於既有全域配準流程；在單執行緒 Intel Core i7-5960X 上，合成距離影像每對平均 0.22 s，比最快的既有全域方法 CZK 快約 50 倍；此方法並可延伸為多片點雲聯合配準。","FGR optimizes a robust Geman-McClure objective over FPFH correspondences with graduated non-convexity, aligning partially overlapping surfaces without initialization.","full_text_reviewed","peer_reviewed_published","background","not_reported",[20,21],"simulation","public_benchmark",[23,24,25,26,27,28],"matches or exceeds accuracy of global pipelines while being at least an order of magnitude faster (abstract)","no initialization required (abstract)","Most robust to noise on synthetic pairs: average RMSE 0.008 and maximal 0.017 at sigma 0.005, against the best prior 0.017 and 0.095 (Table 1)","About 50 times faster than the fastest prior global method CZK (0.22 s vs 11.1 s, Table 2) and faster than PCL ICP and Sparse ICP (Table 3)","Highest precision (23.2%) on the Choi et al. scene benchmark (Table 5); 0.05-recall 84% on UWA (Sec. 5.1)","Multi-way variant matches Choi et al. accuracy (0.05 m average) while about 60 times faster (Table 6)",[30,31,32,33],"depends on FPFH correspondence quality (Sec. 3.3; dependency noted, not an author-stated limitation)","Lower recall than CZK on the Choi et al. scene benchmark (51.1% vs 59.2%, Table 5)","Multi-way accuracy equals but does not exceed Choi et al. (0.05 m average; 0.05 vs 0.04 m on Living room 1 and Office 2, Table 6)","No author-stated limitations; evaluation limited to synthetic and benchmark data without a physical sensor platform (Sec. 5, Sec. 6)",[35,36,37],"synthetic range images (AIM@SHAPE Bimba, Dancing Children, Chinese Dragon; Berkeley Angel; Stanford Bunny) with added 3D Gaussian noise","UWA object and scene benchmark data (acquisition sensor not stated in the paper)","Choi et al. scene fragments whose high-frequency noise and low-frequency distortion simulate consumer depth camera scans; Augmented ICL-NUIM sequences (living room and office) used for multi-way registration",[39],"offline evaluation on synthetic data and public benchmarks; no physical platform","scaled Geman-McClure robust objective optimized via Black-Rangarajan line-process duality with alternating updates and graduated non-convexity (Sec. 3.1-3.2); pose step is a Gauss-Newton solve on a locally linearized 6-vector mapped back to SE(3); mu starts at D^2 (largest surface diameter) and is halved every four iterations down to delta^2; alignment is validated once after convergence (Sec. 3.2, Algorithm 1)","FPFH nearest neighbours in feature space in both directions, filtered by a reciprocity test and a tuple test on three random pairs (edge-length ratios within tau = 0.9 and 1\u002Ftau); correspondences stay fixed during optimization (Sec. 3.3, Algorithm 1)","not_applicable","none","multi-way joint registration: one objective over all fragment poses with robust terms on pairwise correspondences and L2 odometry terms (lambda = 2 in experiments), solved by alternating line-process and 6|T|-variable Gauss-Newton updates without intermediate pairwise alignments (Sec. 4, Sec. 5.2)","surfaces \u002F point clouds","none for pairwise registration; the multi-way variant incorporates initial odometry transformations between consecutive fragments as L2 backbone terms","rigid transformation(s)","Pairwise experiments (Sec. 5.1), single thread on an Intel Core i7-5960X at 3.00 GHz: 0.22 s average per synthetic pair (Table 2), 0.5 s per UWA test (Table 4), 0.2 s per Choi benchmark pair (Table 5); most time is spent on FPFH computation and correspondence building, optimization stays below 30 ms for more than 20,000 points, and validation takes about 3.3% of runtime (Sec. 5.1). Multi-way registration: 82 s average per Augmented ICL-NUIM sequence (Table 6); Sec. 5.2 does not restate the hardware","https:\u002F\u002Fgithub.com\u002Fisl-org\u002FFastGlobalRegistration","MIT (LICENSE file checked)",[52],{"relation":53,"title":54,"doi_or_url":49},"code_release","FastGlobalRegistration",{"id":5,"kind":56,"shortName":7,"title":8,"authors":57,"year":9,"venue":61,"venueType":62,"publisher":63,"volumeIssuePages":64,"doi":65,"arxivId":66,"url":67,"firstPublicDate":68,"publicationStatus":16,"metadataStatus":69,"fulltextStatus":15,"era":10,"classicReason":70,"codeUrl":49,"cluster":11,"topics":71,"mdpi":72,"verification":73,"label":6,"fulltextRoute":74,"versionRead":75,"addedByCensus":72},"method",[58,59,60],"Qian-Yi Zhou","Jaesik Park","Vladlen Koltun","Computer Vision – ECCV 2016 (Lecture Notes in Computer Science)","conference","Springer","pp. 766-782","10.1007\u002F978-3-319-46475-6_47",null,"http:\u002F\u002Fvladlen.info\u002Fpapers\u002Ffast-global-registration.pdf","2016-09-17","metadata_verified","reproducible baseline and principle reused: robust Geman-McClure objective with graduated non-convexity over FPFH matches; first public 2016-09-17, eight days before the recent window, so admitted via the classic route.",[11],false,"corrected","NTU institutional (curl)","Version of record: ECCV 2016 Part II, LNCS 9906, pp. 766-782, © Springer International Publishing AG 2016 (17-page chapter PDF from Springer Link); the author PDF on vladlen.info (16 pages) was also read and its numbers are identical to the version of record.",[77],{"category":78,"model":79,"canonical":79,"role":80,"dataset":66,"specs":81,"locator":82},"compute","Intel Core i7-5960X","compute for runtime","3.00 GHz; all execution times measured with a single thread","Sec. 5.1",[],{"totalRows":85,"groupCount":86,"groups":87,"others":716},53,11,[88,170,411,600],{"slug":89,"group":90,"sourceId":5,"sourceLabel":6,"table":91,"selfRows":92,"metrics":93,"seqs":102,"entrants":115,"cells":121,"outcomes":163,"locators":164,"hardware":165,"wordings":167,"notes":168},"zhou2016fgr-table-6","zhou2016fgr:Table 6","Table 6",10,[94,98,101],{"label":95,"unit":96,"statistic":97,"alignment":43},"Mean error (meters) of reconstructed surface to ground-truth model","m","mean",{"label":99,"unit":100,"statistic":18,"alignment":43},"Time (seconds) for registering all fragments of the sequence","s",{"label":99,"unit":100,"statistic":97,"alignment":43},[103,107,109,111,113],{"dataset":104,"sequence":105,"environment":106},"Augmented ICL-NUIM","Living room 1","indoor scenes (living room and office sequences)",{"dataset":104,"sequence":108,"environment":106},"Living room 2",{"dataset":104,"sequence":110,"environment":106},"Office 1",{"dataset":104,"sequence":112,"environment":106},"Office 2",{"dataset":104,"sequence":114,"environment":106},"Average",[116,118],{"name":117,"methodId":66,"linkable":72,"proposed":72,"self":72},"Choi et al. [7]",{"name":119,"methodId":5,"linkable":120,"proposed":120,"self":120},"Ours (FGR multi-way)",true,[122,126,129,131,133,136,137,139,140,142,143,145,147,149,151,153,155,157,159,161],[123,123,123,124,125,123,125,125,123],0,0.04,-1,[127,123,123,128,125,123,125,125,123],1,0.05,[123,123,127,130,125,123,125,125,123],0.07,[127,123,127,132,125,123,125,125,123],0.06,[123,123,134,135,125,123,125,125,123],2,0.03,[127,123,134,135,125,123,125,125,123],[123,123,138,124,125,123,125,125,123],3,[127,123,138,128,125,123,125,125,123],[123,123,141,128,125,123,125,125,123],4,[127,123,141,128,125,123,125,125,123],[123,127,123,144,125,123,123,125,123],8940,[127,127,123,146,125,123,123,125,123],131,[123,127,127,148,125,123,123,125,123],3360,[127,127,127,150,125,123,123,125,123],81,[123,127,134,152,125,123,123,125,123],4500,[127,127,134,154,125,123,123,125,123],69,[123,127,138,156,125,123,123,125,123],4080,[127,127,138,158,125,123,123,125,123],48,[123,134,141,160,125,123,123,125,123],5220,[127,134,141,162,125,123,123,125,123],82,[],[91],[166],"not stated for Sec. 5.2 (Sec. 5.1 states Intel Core i7-5960X single thread)",[],[169],"Multi-way registration of all fragments (47 to 57 per sequence), lambda = 2; mean distance of integrated surface to ground-truth model and total time",{"slug":171,"group":172,"sourceId":173,"sourceLabel":174,"table":91,"selfRows":175,"metrics":176,"seqs":183,"entrants":192,"cells":231,"outcomes":404,"locators":406,"hardware":407,"wordings":408,"notes":409},"lim2024quatropp-table-6","lim2024quatropp:Table 6","lim2024quatropp","Lim et al., 2024",6,[177,180],{"label":178,"unit":179,"statistic":18,"alignment":18},"trel","%",{"label":181,"unit":182,"statistic":18,"alignment":18},"rrel","deg\u002F100m",[184,188,190],{"dataset":185,"sequence":186,"environment":187},"KITTI","Seq. 00, Delta = 1","vehicle, urban driving (Velodyne HDL-64E)",{"dataset":185,"sequence":189,"environment":187},"Seq. 00, Delta = 3",{"dataset":185,"sequence":191,"environment":187},"Seq. 00, Delta = 5",[193,196,199,202,203,206,208,210,213,215,217,219,221,224,227,229],{"name":194,"methodId":195,"linkable":120,"proposed":72,"self":72},"ICP","besl1992icp",{"name":197,"methodId":198,"linkable":120,"proposed":72,"self":72},"G-ICP","segal2009gicp",{"name":200,"methodId":201,"linkable":120,"proposed":72,"self":72},"VGICP","koide2021vgicp",{"name":7,"methodId":5,"linkable":120,"proposed":72,"self":120},{"name":204,"methodId":205,"linkable":120,"proposed":72,"self":72},"TEASER++","yang2021teaser",{"name":207,"methodId":66,"linkable":72,"proposed":120,"self":72},"Quatro (Ours)",{"name":209,"methodId":173,"linkable":120,"proposed":120,"self":72},"Quatro++ (Ours)",{"name":211,"methodId":212,"linkable":120,"proposed":72,"self":72},"LO-Net","lonet2019",{"name":214,"methodId":212,"linkable":120,"proposed":72,"self":72},"LO-Net+M",{"name":216,"methodId":66,"linkable":72,"proposed":72,"self":72},"DMLO†",{"name":218,"methodId":66,"linkable":72,"proposed":72,"self":72},"DMLO+M†",{"name":220,"methodId":66,"linkable":72,"proposed":72,"self":72},"A-LOAM + StickyPillars†",{"name":222,"methodId":223,"linkable":120,"proposed":72,"self":72},"SuMa","suma2018",{"name":225,"methodId":226,"linkable":120,"proposed":72,"self":72},"A-LOAM","aloam_software",{"name":228,"methodId":66,"linkable":72,"proposed":120,"self":72},"Quatro-c2f (Ours)",{"name":230,"methodId":173,"linkable":120,"proposed":120,"self":72},"Quatro++-c2f (Ours)",[232,234,236,238,240,242,244,246,248,250,252,254,256,258,260,262,264,266,268,270,272,274,276,278,280,282,284,286,288,290,291,293,295,297,299,301,303,305,307,308,310,312,314,317,319,320,321,322,323,326,328,329,330,331,332,335,337,338,339,340,341,343,344,345,346,347,348,350,352,354,356,358,360,363,365,367,369,371,373,376,378,380,382,384,386,388,390,392,393,394,395,397,398,399,400,402],[123,123,123,233,125,123,125,125,123],6.88,[123,127,123,235,125,123,125,125,123],2.99,[123,123,127,237,125,123,125,125,123],21.92,[123,127,127,239,125,123,125,125,123],8.7,[123,123,134,241,125,123,125,125,123],21.14,[123,127,134,243,125,123,125,125,123],8.51,[127,123,123,245,125,123,125,125,123],1.26,[127,127,123,247,125,123,125,125,123],0.45,[127,123,127,249,125,123,125,125,123],5.5,[127,127,127,251,125,123,125,125,123],1.45,[127,123,134,253,125,123,125,125,123],14.2,[127,127,134,255,125,123,125,125,123],3.32,[134,123,123,257,125,123,125,125,123],1.03,[134,127,123,259,125,123,125,125,123],0.3,[134,123,127,261,125,123,125,125,123],11.83,[134,127,127,263,125,123,125,125,123],1.65,[134,123,134,265,125,123,125,125,123],19.11,[134,127,134,267,125,123,125,125,123],6.32,[138,123,123,269,125,123,125,125,123],2.73,[138,127,123,271,125,123,125,125,123],0.69,[138,123,127,273,125,123,125,125,123],7.17,[138,127,127,275,125,123,125,125,123],1.58,[138,123,134,277,125,123,125,125,123],14.66,[138,127,134,279,125,123,125,125,123],4.12,[141,123,123,281,125,123,125,125,123],2.11,[141,127,123,283,125,123,125,125,123],0.91,[141,123,127,285,125,123,125,125,123],2.64,[141,127,127,287,125,123,125,125,123],1.11,[141,123,134,289,125,123,125,125,123],3.19,[141,127,134,283,125,123,125,125,123],[292,123,123,251,125,123,125,125,123],5,[292,127,123,294,125,123,125,125,123],0.41,[292,123,127,296,125,123,125,125,123],1.38,[292,127,127,298,125,123,125,125,123],0.24,[292,123,134,300,125,123,125,125,123],1.94,[292,127,134,302,125,123,125,125,123],0.46,[175,123,123,304,125,123,125,125,123],1.9,[175,127,123,306,125,123,125,125,123],0.53,[175,123,127,251,125,123,125,125,123],[175,127,127,309,125,123,125,125,123],0.32,[175,123,134,311,125,123,125,125,123],0.99,[175,127,134,313,125,123,125,125,123],0.28,[315,123,123,316,125,123,125,125,123],7,1.47,[315,127,123,318,125,123,125,125,123],0.72,[315,123,127,66,123,123,125,125,123],[315,127,127,66,123,123,125,125,123],[315,123,134,66,123,123,125,125,123],[315,127,134,66,123,123,125,125,123],[324,123,123,325,125,123,125,125,123],8,0.78,[324,127,123,327,125,123,125,125,123],0.42,[324,123,127,66,123,123,125,125,123],[324,127,127,66,123,123,125,125,123],[324,123,134,66,123,123,125,125,123],[324,127,134,66,123,123,125,125,123],[333,123,123,334,125,123,125,125,123],9,0.83,[333,127,123,336,125,123,125,125,123],0.44,[333,123,127,66,123,123,125,125,123],[333,127,127,66,123,123,125,125,123],[333,123,134,66,123,123,125,125,123],[333,127,134,66,123,123,125,125,123],[92,123,123,342,125,123,125,125,123],0.73,[92,127,123,336,125,123,125,125,123],[92,123,127,66,123,123,125,125,123],[92,127,127,66,123,123,125,125,123],[92,123,134,66,123,123,125,125,123],[92,127,134,66,123,123,125,125,123],[86,123,123,349,125,123,125,125,123],0.65,[86,127,123,351,125,123,125,125,123],0.26,[86,123,127,353,125,123,125,125,123],0.79,[86,127,127,355,125,123,125,125,123],0.31,[86,123,134,357,125,123,125,125,123],1.29,[86,127,134,359,125,123,125,125,123],0.48,[361,123,123,362,125,123,125,125,123],12,0.68,[361,127,123,364,125,123,125,125,123],0.23,[361,123,127,366,125,123,125,125,123],1.69,[361,127,127,368,125,123,125,125,123],0.61,[361,123,134,370,125,123,125,125,123],2.36,[361,127,134,372,125,123,125,125,123],0.51,[374,123,123,375,125,123,125,125,123],13,0.7,[374,127,123,377,125,123,125,125,123],0.27,[374,123,127,379,125,123,125,125,123],0.97,[374,127,127,381,125,123,125,125,123],0.38,[374,123,134,383,125,123,125,125,123],31.16,[374,127,134,385,125,123,125,125,123],12.1,[387,123,123,349,125,123,125,125,123],14,[387,127,123,389,125,123,125,125,123],0.21,[387,123,127,391,125,123,125,125,123],0.67,[387,127,127,389,125,123,125,125,123],[387,123,134,391,125,123,125,125,123],[387,127,134,389,125,123,125,125,123],[396,123,123,362,125,123,125,125,123],15,[396,127,123,364,125,123,125,125,123],[396,123,127,368,125,123,125,125,123],[396,127,127,389,125,123,125,125,123],[396,123,134,401,125,123,125,125,123],0.5,[396,127,134,403,125,123,125,125,123],0.2,[405],"not_reported (N\u002FA in table; not available from the original paper)",[91],[],[],[410],"KITTI Seq. 00 odometry test with frame interval Delta (source i+Delta, target i); trel [%] and rrel [deg\u002F100m] by RPG evaluation tools; c2f = global registration then local registration (G-ICP); deep-learning rows copied by the authors from the original papers; † = Seq. 00 used for training",{"slug":412,"group":413,"sourceId":414,"sourceLabel":415,"table":416,"selfRows":175,"metrics":417,"seqs":432,"entrants":437,"cells":461,"outcomes":594,"locators":595,"hardware":596,"wordings":597,"notes":598},"zhang2024globalbimreg-table-1","zhang2024globalbimreg:Table 1","zhang2024globalbimreg","Zhang et al., 2024b","Table 1",[418,422,424,426,428,430],{"label":419,"unit":420,"statistic":421,"alignment":43},"RE_50, rotation error 50th percentile","deg","median",{"label":423,"unit":420,"statistic":18,"alignment":43},"RE_75, rotation error 75th percentile",{"label":425,"unit":420,"statistic":18,"alignment":43},"RE_95, rotation error 95th percentile",{"label":427,"unit":96,"statistic":421,"alignment":43},"TE_50, translation error 50th percentile",{"label":429,"unit":96,"statistic":18,"alignment":43},"TE_75, translation error 75th percentile",{"label":431,"unit":96,"statistic":18,"alignment":43},"TE_95, translation error 95th percentile",[433],{"dataset":434,"sequence":435,"environment":436},"ISPRS benchmark on indoor modelling","Models 01-05 (250 samples)","building interiors from the ISPRS indoor modelling benchmark (Models 01-05)",[438,440,442,445,447,448,451,453,455,457,459],{"name":439,"methodId":66,"linkable":72,"proposed":72,"self":72},"GMMTree (initialised with FPFH-RANSAC)",{"name":441,"methodId":66,"linkable":72,"proposed":72,"self":72},"FilterReg (initialised with FPFH-RANSAC)",{"name":443,"methodId":444,"linkable":120,"proposed":72,"self":72},"GO-ICP","yang2016goicp",{"name":446,"methodId":66,"linkable":72,"proposed":72,"self":72},"Super4PCS",{"name":7,"methodId":5,"linkable":120,"proposed":72,"self":120},{"name":449,"methodId":450,"linkable":120,"proposed":72,"self":72},"RANSAC (FPFH features)","fischler1981ransac",{"name":452,"methodId":66,"linkable":72,"proposed":72,"self":72},"RMMG",{"name":454,"methodId":66,"linkable":72,"proposed":72,"self":72},"PLADE",{"name":456,"methodId":66,"linkable":72,"proposed":72,"self":72},"DCP",{"name":458,"methodId":66,"linkable":72,"proposed":72,"self":72},"PointNetLK",{"name":460,"methodId":414,"linkable":120,"proposed":120,"self":72},"Ours (primitive-level coarse registration)",[462,464,466,468,470,472,474,476,478,480,482,484,486,488,490,492,494,496,498,500,502,504,506,508,510,512,514,516,518,520,522,524,526,528,530,532,534,536,538,540,542,544,546,548,550,552,554,556,558,560,562,564,566,568,570,572,574,576,578,580,582,584,586,588,590,592],[123,123,123,463,125,123,125,125,123],4.466,[123,127,123,465,125,123,125,125,123],8.214,[123,134,123,467,125,123,125,125,123],19.348,[123,138,123,469,125,123,125,125,123],1.652,[123,141,123,471,125,123,125,125,123],3.166,[123,292,123,473,125,123,125,125,123],5.419,[127,123,123,475,125,123,125,125,123],3.59,[127,127,123,477,125,123,125,125,123],10.906,[127,134,123,479,125,123,125,125,123],18.869,[127,138,123,481,125,123,125,125,123],1.076,[127,141,123,483,125,123,125,125,123],1.768,[127,292,123,485,125,123,125,125,123],5.279,[134,123,123,487,125,123,125,125,123],4.178,[134,127,123,489,125,123,125,125,123],4.252,[134,134,123,491,125,123,125,125,123],10.618,[134,138,123,493,125,123,125,125,123],1.787,[134,141,123,495,125,123,125,125,123],2.398,[134,292,123,497,125,123,125,125,123],5.216,[138,123,123,499,125,123,125,125,123],2.208,[138,127,123,501,125,123,125,125,123],5.162,[138,134,123,503,125,123,125,125,123],10.693,[138,138,123,505,125,123,125,125,123],1.14,[138,141,123,507,125,123,125,125,123],2.31,[138,292,123,509,125,123,125,125,123],5.591,[141,123,123,511,125,123,125,125,123],27.139,[141,127,123,513,125,123,125,125,123],35.195,[141,134,123,515,125,123,125,125,123],44.235,[141,138,123,517,125,123,125,125,123],1.969,[141,141,123,519,125,123,125,125,123],3.674,[141,292,123,521,125,123,125,125,123],5.094,[292,123,123,523,125,123,125,125,123],5.477,[292,127,123,525,125,123,125,125,123],7.912,[292,134,123,527,125,123,125,125,123],13.7,[292,138,123,529,125,123,125,125,123],1.412,[292,141,123,531,125,123,125,125,123],2.032,[292,292,123,533,125,123,125,125,123],5.253,[175,123,123,535,125,123,125,125,123],1.673,[175,127,123,537,125,123,125,125,123],3.351,[175,134,123,539,125,123,125,125,123],5.669,[175,138,123,541,125,123,125,125,123],1.335,[175,141,123,543,125,123,125,125,123],1.983,[175,292,123,545,125,123,125,125,123],8.136,[315,123,123,547,125,123,125,125,123],0.424,[315,127,123,549,125,123,125,125,123],0.623,[315,134,123,551,125,123,125,125,123],1.473,[315,138,123,553,125,123,125,125,123],3.83,[315,141,123,555,125,123,125,125,123],6.554,[315,292,123,557,125,123,125,125,123],9.659,[324,123,123,559,125,123,125,125,123],34.865,[324,127,123,561,125,123,125,125,123],43.069,[324,134,123,563,125,123,125,125,123],44.88,[324,138,123,565,125,123,125,125,123],1.945,[324,141,123,567,125,123,125,125,123],2.667,[324,292,123,569,125,123,125,125,123],9.945,[333,123,123,571,125,123,125,125,123],11.741,[333,127,123,573,125,123,125,125,123],21.618,[333,134,123,575,125,123,125,125,123],40.482,[333,138,123,577,125,123,125,125,123],5.007,[333,141,123,579,125,123,125,125,123],6.566,[333,292,123,581,125,123,125,125,123],9.231,[92,123,123,583,125,123,125,125,123],0.272,[92,127,123,585,125,123,125,125,123],0.329,[92,134,123,587,125,123,125,125,123],0.409,[92,138,123,589,125,123,125,125,123],0.053,[92,141,123,591,125,123,125,125,123],0.148,[92,292,123,593,125,123,125,125,123],0.352,[],[416],[],[],[599],"Coarse registration on 250 samples (50 per model) with random rigid perturbations (roll and pitch within 30 deg, yaw within 180 deg, translation within 10 m); only successful results with RE \u003C 45 deg and TE \u003C 10 m are included; A50, A75 and A95 quantiles of rotation and translation error against the benchmark alignment",{"slug":601,"group":602,"sourceId":5,"sourceLabel":6,"table":416,"selfRows":175,"metrics":603,"seqs":610,"entrants":619,"cells":635,"outcomes":710,"locators":711,"hardware":712,"wordings":713,"notes":714},"zhou2016fgr-table-1","zhou2016fgr:Table 1",[604,607],{"label":605,"unit":606,"statistic":97,"alignment":43},"Average RMSE","surface diameter",{"label":608,"unit":606,"statistic":609,"alignment":43},"Maximal RMSE","max",[611,615,617],{"dataset":612,"sequence":613,"environment":614},"Synthetic range images (AIM@SHAPE, Berkeley Angel, Stanford Bunny)","sigma 0","synthetic",{"dataset":612,"sequence":616,"environment":614},"sigma 0.0025",{"dataset":612,"sequence":618,"environment":614},"sigma 0.005",[620,622,624,626,628,631,633],{"name":621,"methodId":444,"linkable":120,"proposed":72,"self":72},"GoICP [42]",{"name":623,"methodId":444,"linkable":120,"proposed":72,"self":72},"GoICP-Trimming [42]",{"name":625,"methodId":66,"linkable":72,"proposed":72,"self":72},"Super 4PCS [26]",{"name":627,"methodId":66,"linkable":72,"proposed":72,"self":72},"OpenCV [8] (implementation of Drost et al.)",{"name":629,"methodId":630,"linkable":120,"proposed":72,"self":72},"PCL [19,34] (PCL implementation of Rusu et al.)","rusu2009fpfh",{"name":632,"methodId":66,"linkable":72,"proposed":72,"self":72},"CZK [7] (Choi et al. variant of Rusu's algorithm)",{"name":634,"methodId":5,"linkable":120,"proposed":120,"self":120},"Our approach (FGR)",[636,638,640,642,644,646,648,650,652,654,656,658,660,662,664,666,667,669,671,673,675,677,679,680,682,684,686,687,689,691,693,694,695,697,699,700,702,703,704,706,708,709],[123,123,123,637,125,123,125,125,123],0.029,[123,127,123,639,125,123,125,125,123],0.13,[123,123,127,641,125,123,125,125,123],0.032,[123,127,127,643,125,123,125,125,123],0.133,[123,123,134,645,125,123,125,125,123],0.037,[123,127,134,647,125,123,125,125,123],0.127,[127,123,123,649,125,123,125,125,123],0.035,[127,127,123,651,125,123,125,125,123],0.473,[127,123,127,653,125,123,125,125,123],0.039,[127,127,127,655,125,123,125,125,123],0.475,[127,123,134,657,125,123,125,125,123],0.044,[127,127,134,659,125,123,125,125,123],0.478,[134,123,123,661,125,123,125,125,123],0.012,[134,127,123,663,125,123,125,125,123],0.019,[134,123,127,665,125,123,125,125,123],0.014,[134,127,127,637,125,123,125,125,123],[134,123,134,668,125,123,125,125,123],0.017,[134,127,134,670,125,123,125,125,123],0.095,[138,123,123,672,125,123,125,125,123],0.009,[138,127,123,674,125,123,125,125,123],0.013,[138,123,127,676,125,123,125,125,123],0.018,[138,127,127,678,125,123,125,125,123],0.212,[138,123,134,641,125,123,125,125,123],[138,127,134,681,125,123,125,125,123],0.242,[141,123,123,683,125,123,125,125,123],0.003,[141,127,123,685,125,123,125,125,123],0.005,[141,123,127,672,125,123,125,125,123],[141,127,127,688,125,123,125,125,123],0.061,[141,123,134,690,125,123,125,125,123],0.111,[141,127,134,692,125,123,125,125,123],0.414,[292,123,123,683,125,123,125,125,123],[292,127,123,685,125,123,125,125,123],[292,123,127,696,125,123,125,125,123],0.008,[292,127,127,698,125,123,125,125,123],0.022,[292,123,134,649,125,123,125,125,123],[292,127,134,701,125,123,125,125,123],0.274,[175,123,123,683,125,123,125,125,123],[175,127,123,685,125,123,125,125,123],[175,123,127,705,125,123,125,125,123],0.006,[175,127,127,707,125,123,125,125,123],0.011,[175,123,134,696,125,123,125,125,123],[175,127,134,668,125,123,125,125,123],[],[416],[],[],[715],"25 synthetic range-image pairs per noise level; RMSE of ground-truth correspondence distances, unit surface diameter; GoICP variants on 1,000 points",[717,723,728,734,740,745,751],{"group":718,"slug":719,"sourceLabel":6,"table":720,"selfRows":175,"datasets":721},"zhou2016fgr:Table 2","zhou2016fgr-table-2","Table 2",[722],"Synthetic range images",{"group":724,"slug":725,"sourceLabel":6,"table":726,"selfRows":175,"datasets":727},"zhou2016fgr:Table 3","zhou2016fgr-table-3","Table 3",[722],{"group":729,"slug":730,"sourceLabel":731,"table":732,"selfRows":292,"datasets":733},"qin2023geotransformer:Table 5","qin2023geotransformer-table-5","Qin et al., 2023","Table 5",[104],{"group":735,"slug":736,"sourceLabel":737,"table":738,"selfRows":138,"datasets":739},"lim2025kissmatcher:Table I","lim2025kissmatcher-table-i","Lim et al., 2025","Table I",[185],{"group":741,"slug":742,"sourceLabel":6,"table":732,"selfRows":138,"datasets":743},"zhou2016fgr:Table 5","zhou2016fgr-table-5",[744],"Choi et al. scene benchmark",{"group":746,"slug":747,"sourceLabel":6,"table":748,"selfRows":127,"datasets":749},"zhou2016fgr:Table 4","zhou2016fgr-table-4","Table 4",[750],"UWA benchmark",{"group":752,"slug":753,"sourceLabel":6,"table":754,"selfRows":127,"datasets":755},"zhou2016fgr:Text Sec.5.1","zhou2016fgr-text-sec-5-1","Text Sec.5.1",[750],1790510663935]