[{"data":1,"prerenderedAt":587},["ShallowReactive",2],{"method-yang2016goicp":3},{"method":4,"reference":55,"equipment":77,"figures":97,"results":98},{"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":24,"sensors":32,"platform":34,"estimator":35,"association":36,"timeModel":37,"deskew":37,"loopClosure":38,"globalOptimization":39,"mapRepresentation":40,"prior":41,"outputGeometry":42,"compute":43,"codeUrl":44,"codeLicense":45,"relatedVersions":46},"yang2016goicp","Yang et al., 2016","Go-ICP","Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration",2016,"classic","C02","registration_component","Go-ICP 在整個 SE(3) 空間以分支定界（BnB）搜尋點對點 ICP 之 L2 誤差的全域最佳解。旋轉以角軸向量表示於 [-π, π]³ 立方體，平移限定在 [-ξ, ξ]³，兩者都以八元樹細分；作者由旋轉與平移的不確定半徑推導每點殘差的上下界，並採外層旋轉、內層平移的巢狀 BnB。每當找到更好的解就以局部 ICP 精修並更新上界，加快收斂而不失全域最佳性；另以修剪（trimming）處理部分重疊與離群點。在 Stanford bunny 與 dragon 的 2000 次部分對完整配準中全部成功，使用距離轉換時平均約 1.5 至 1.6 秒、最長 28.9 秒（Intel i7 3.4 GHz，1000 個資料點）。作者建議用於不要求即時性的情境，或作為最佳性基準。","Branch-and-bound over SE(3) with derived bounds gives globally optimal L2 ICP registration independent of initialization.","full_text_reviewed","peer_reviewed_published","supplementary","not_reported（僅有物件模型、辦公室場景深度影像定位與 RGB-D 外參校正示例，無營建場域）",[20,21],"simulation","public_benchmark",[23],"globally optimal regardless of initialization (abstract, Sec. 7); 100% correct registration on 2,000 partial-to-full tasks on bunny and dragon, rotation errors \u003C 2 deg and translation errors \u003C 0.01 in normalized units (Sec. 6.2); all 2,000 trimmed partial-overlap tasks on 10 point-set pairs with 50% to 95% overlap correct, rotation \u003C 5 deg and translation \u003C 0.05 (Sec. 6.3, Table 1); camera localization of 100 depth images against an office model with errors below 5 deg and 10 cm (Sec. 6.4)",[25,26,27,28,29,30,31],"runs take seconds to minutes, and the authors recommend it where real-time performance is not critical or as an optimality benchmark (Sec. 6, 7)","experiments sub-sample data sets to 1000 (or 400 to 600) data points (Sec. 6.2 to 6.4)","runtime grows when the optimal RMS error is high, e.g. global minimum found at about 25 s with the rest spent raising the lower bound (Sec. 6.2, Fig. 17)","trimming percentage chosen by visually guessing the non-overlap ratio (Sec. 6.3)","L2 objective is outlier-sensitive and only trimming is implemented (Sec. 2, 5.3)","effective data size is limited (bueno2018plcs Sec. 2, secondary)","in FGR's comparison, Go-ICP and Go-ICP-Trimming inputs were downsampled to 1,000 points as suggested by Yang et al. (zhou2016fgr Sec. 5, secondary)",[33],"[\"Kinect (bowl and loom point sets)\", \"structured light 3D scanner (denture point set)\", \"RGB-D depth images from public datasets (camera localization dataset [68], RGB-D Object Dataset [69])\"]",[],"nested best-first branch-and-bound: an outer BnB over rotation (angle-axis cube [-pi, pi]^3 split by octree) calls an inner BnB over translation (cube [-xi, xi]^3); per-point residual bounds come from rotation and translation uncertainty radii; whenever a better cube is found, local ICP is run from it and its result tightens the upper bound; the search stops when the best error minus the lower bound is below epsilon; outliers handled with a trimmed L2 error using Introselect (O(N))","closest point under the L2 residual; for bound evaluation closest distances come either from a kd-tree or, more often in the experiments, from a precomputed 3-D Euclidean distance transform (300 x 300 x 300 grid, approximate); local ICP always uses a kd-tree","not_applicable","none","globally optimal search over SE(3) (abstract)","3D point sets","no initial pose needed; requires a bounded translation domain [-xi, xi]^3, point sets pre-normalized to [-1, 1]^3, a convergence threshold epsilon and, for partial overlap, a hand-chosen trimming percentage rho","rigid transformation","C++ on a standard PC with an Intel i7 3.4 GHz CPU; with the distance transform, mean and longest times 1.6 s and 22.3 s (bunny) and 1.5 s and 28.9 s (dragon) for 1000 data points against 20,000 to 40,000 model points; kd-tree runs typically 40 to 50 times longer; trimmed partial-overlap runs 0.45 to 18.4 s mean and up to 107.3 s max (Table 1); camera localization 32 s mean and 178 s max","https:\u002F\u002Fgithub.com\u002Fyangjiaolong\u002FGo-ICP","GPL-3.0 (LICENSE file checked)",[47,51],{"relation":48,"title":49,"doi_or_url":50},"conference_version","Go-ICP: Solving 3D Registration Efficiently and Globally Optimally (ICCV 2013)","10.1109\u002FICCV.2013.184",{"relation":52,"title":53,"doi_or_url":54},"preprint","arXiv:1605.03344 v1 (2016-05-11), posted after the TPAMI online date","https:\u002F\u002Farxiv.org\u002Fabs\u002F1605.03344",{"id":5,"kind":56,"shortName":7,"title":8,"authors":57,"year":9,"venue":62,"venueType":63,"publisher":64,"volumeIssuePages":65,"doi":66,"arxivId":67,"url":68,"firstPublicDate":69,"publicationStatus":16,"metadataStatus":70,"fulltextStatus":15,"era":10,"classicReason":71,"codeUrl":44,"cluster":11,"topics":72,"mdpi":73,"verification":74,"label":6,"fulltextRoute":75,"versionRead":76,"addedByCensus":73},"method",[58,59,60,61],"Jiaolong Yang","Hongdong Li","Dylan Campbell","Yunde Jia","IEEE Transactions on Pattern Analysis and Machine Intelligence","journal","IEEE","38(11):2241-2254","10.1109\u002Ftpami.2015.2513405","1605.03344","https:\u002F\u002Fapi.crossref.org\u002Fworks?query.bibliographic=Go-ICP%3A%20A%20Globally%20Optimal%20Solution%20to%203D%20ICP%20Point-Set%20Registration","2015-12-30","metadata_verified","reproducible baseline: branch-and-bound globally optimal ICP used as comparison baseline by FGR and TEASER++.",[11],false,"corrected","arXiv","arXiv 1605.03344v1 (11 May 2016), 14 pages; posted after the TPAMI online date, textual equivalence to the TPAMI version of record not checked; supplementary material not read",[78,85,92],{"category":79,"model":80,"canonical":80,"role":81,"dataset":82,"specs":83,"locator":84},"compute","Intel i7 3.4GHz CPU (standard PC)","compute for runtime",null,"C++ implementation","Sec. 6",{"category":86,"model":87,"canonical":87,"role":88,"dataset":89,"specs":90,"locator":91},"rgbd","Kinect","dataset sensor","bowl and loom point sets collected by the authors","not_reported","Sec. 6.3",{"category":93,"model":94,"canonical":94,"role":88,"dataset":95,"specs":90,"locator":96},"other","structured light 3D scanner","denture point set","Sec. 6.3, footnote 6",[],{"totalRows":99,"groupCount":100,"groups":101,"others":555},87,9,[102,268,385,500],{"slug":103,"group":104,"sourceId":5,"sourceLabel":6,"table":105,"selfRows":106,"metrics":107,"seqs":114,"entrants":149,"cells":159,"outcomes":260,"locators":261,"hardware":262,"wordings":264,"notes":265},"yang2016goicp-table-1","yang2016goicp:Table 1","Table 1",40,[108,112],{"label":109,"unit":110,"statistic":111,"alignment":38},"mean\u002Fmax time (s)","s","mean",{"label":109,"unit":110,"statistic":113,"alignment":38},"max",[115,119,121,123,124,126,127,129,130,132,133,135,136,138,139,142,143,145,146,148],{"dataset":116,"sequence":117,"environment":118},"Bunny (Stanford 3D)","A to B","object scans",{"dataset":116,"sequence":120,"environment":118},"B to A",{"dataset":122,"sequence":117,"environment":118},"Dragon (Stanford 3D)",{"dataset":122,"sequence":120,"environment":118},{"dataset":125,"sequence":117,"environment":118},"Buddha (Stanford 3D)",{"dataset":125,"sequence":120,"environment":118},{"dataset":128,"sequence":117,"environment":118},"Chef (ref. [67])",{"dataset":128,"sequence":120,"environment":118},{"dataset":131,"sequence":117,"environment":118},"Dinosaur (ref. [67])",{"dataset":131,"sequence":120,"environment":118},{"dataset":134,"sequence":117,"environment":118},"Owl (ref. [66])",{"dataset":134,"sequence":120,"environment":118},{"dataset":137,"sequence":117,"environment":118},"Denture (structured light scanner)",{"dataset":137,"sequence":120,"environment":118},{"dataset":140,"sequence":117,"environment":141},"Room (ref. [68])","indoor room scans",{"dataset":140,"sequence":120,"environment":141},{"dataset":144,"sequence":117,"environment":118},"Bowl (Kinect, authors)",{"dataset":144,"sequence":120,"environment":118},{"dataset":147,"sequence":117,"environment":118},"Loom (Kinect, authors)",{"dataset":147,"sequence":120,"environment":118},[150,153,155,157],{"name":151,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT, trimming rho = 10%)",true,{"name":154,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT, trimming rho = 20%)",{"name":156,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT, trimming rho = 40%)",{"name":158,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT, trimming rho = 30%)",[160,164,167,169,171,174,176,179,181,184,186,189,191,194,196,199,201,204,206,208,210,213,215,218,220,223,225,228,230,233,235,238,240,243,245,248,250,253,255,258],[161,161,161,162,163,161,161,163,161],0,0.81,-1,[161,165,161,166,163,161,161,163,165],1,10.7,[161,161,165,168,163,161,161,163,165],0.49,[161,165,165,170,163,161,161,163,165],7.25,[165,161,172,173,163,161,161,163,165],2,2.99,[165,165,172,175,163,161,161,163,165],43.5,[172,161,177,178,163,161,161,163,165],3,8.72,[172,165,177,180,163,161,161,163,165],72.4,[161,161,182,183,163,161,161,163,165],4,0.71,[161,165,182,185,163,161,161,163,165],11.3,[161,161,187,188,163,161,161,163,165],5,0.6,[161,165,187,190,163,161,161,163,165],14.8,[165,161,192,193,163,161,161,163,165],6,0.45,[165,165,192,195,163,161,161,163,165],4.47,[177,161,197,198,163,161,161,163,165],7,0.52,[177,165,197,200,163,161,161,163,165],3.79,[161,161,202,203,163,161,161,163,165],8,2.03,[161,165,202,205,163,161,161,163,165],23.5,[161,161,100,207,163,161,161,163,165],1.65,[161,165,100,209,163,161,161,163,165],26.1,[172,161,211,212,163,161,161,163,165],10,12.5,[172,165,211,214,163,161,161,163,165],87.5,[172,161,216,217,163,161,161,163,165],11,13.4,[172,165,216,219,163,161,161,163,165],75,[177,161,221,222,163,161,161,163,165],12,6.74,[177,165,221,224,163,161,161,163,165],74.7,[177,161,226,227,163,161,161,163,165],13,4.24,[177,165,226,229,163,161,161,163,165],68.1,[177,161,231,232,163,161,161,163,165],14,9.82,[177,165,231,234,163,161,161,163,165],73.3,[177,161,236,237,163,161,161,163,165],15,18.4,[177,165,236,239,163,161,161,163,165],107.3,[165,161,241,242,163,161,161,163,165],16,3.19,[165,165,241,244,163,161,161,163,165],20.3,[177,161,246,247,163,161,161,163,165],17,3.52,[177,165,246,249,163,161,161,163,165],25.3,[177,161,251,252,163,161,161,163,165],18,8.64,[177,165,251,254,163,161,161,163,165],67.2,[165,161,256,257,163,161,161,163,165],19,5.96,[165,165,256,259,163,161,161,163,165],44.6,[],[105],[263],"Intel i7 3.4 GHz PC",[],[266,267],"Go-ICP with distance transform and trimming on 10 partially overlapping point-set pairs; 100 random relative poses per pair and direction; N = 1000 data points; epsilon = 0.001 x K; all tasks registered correctly","Same setting as other Table 1 rows",{"slug":269,"group":270,"sourceId":271,"sourceLabel":272,"table":105,"selfRows":221,"metrics":273,"seqs":279,"entrants":288,"cells":304,"outcomes":379,"locators":380,"hardware":381,"wordings":382,"notes":383},"zhou2016fgr-table-1","zhou2016fgr:Table 1","zhou2016fgr","Zhou et al., 2016",[274,277],{"label":275,"unit":276,"statistic":111,"alignment":38},"Average RMSE","surface diameter",{"label":278,"unit":276,"statistic":113,"alignment":38},"Maximal RMSE",[280,284,286],{"dataset":281,"sequence":282,"environment":283},"Synthetic range images (AIM@SHAPE, Berkeley Angel, Stanford Bunny)","sigma 0","synthetic",{"dataset":281,"sequence":285,"environment":283},"sigma 0.0025",{"dataset":281,"sequence":287,"environment":283},"sigma 0.005",[289,291,293,295,297,300,302],{"name":290,"methodId":5,"linkable":152,"proposed":73,"self":152},"GoICP [42]",{"name":292,"methodId":5,"linkable":152,"proposed":73,"self":152},"GoICP-Trimming [42]",{"name":294,"methodId":82,"linkable":73,"proposed":73,"self":73},"Super 4PCS [26]",{"name":296,"methodId":82,"linkable":73,"proposed":73,"self":73},"OpenCV [8] (implementation of Drost et al.)",{"name":298,"methodId":299,"linkable":152,"proposed":73,"self":73},"PCL [19,34] (PCL implementation of Rusu et al.)","rusu2009fpfh",{"name":301,"methodId":82,"linkable":73,"proposed":73,"self":73},"CZK [7] (Choi et al. variant of Rusu's algorithm)",{"name":303,"methodId":271,"linkable":152,"proposed":152,"self":73},"Our approach (FGR)",[305,307,309,311,313,315,317,319,321,323,325,327,329,331,333,335,336,338,340,342,344,346,348,349,351,353,355,356,358,360,362,363,364,366,368,369,371,372,373,375,377,378],[161,161,161,306,163,161,163,163,161],0.029,[161,165,161,308,163,161,163,163,161],0.13,[161,161,165,310,163,161,163,163,161],0.032,[161,165,165,312,163,161,163,163,161],0.133,[161,161,172,314,163,161,163,163,161],0.037,[161,165,172,316,163,161,163,163,161],0.127,[165,161,161,318,163,161,163,163,161],0.035,[165,165,161,320,163,161,163,163,161],0.473,[165,161,165,322,163,161,163,163,161],0.039,[165,165,165,324,163,161,163,163,161],0.475,[165,161,172,326,163,161,163,163,161],0.044,[165,165,172,328,163,161,163,163,161],0.478,[172,161,161,330,163,161,163,163,161],0.012,[172,165,161,332,163,161,163,163,161],0.019,[172,161,165,334,163,161,163,163,161],0.014,[172,165,165,306,163,161,163,163,161],[172,161,172,337,163,161,163,163,161],0.017,[172,165,172,339,163,161,163,163,161],0.095,[177,161,161,341,163,161,163,163,161],0.009,[177,165,161,343,163,161,163,163,161],0.013,[177,161,165,345,163,161,163,163,161],0.018,[177,165,165,347,163,161,163,163,161],0.212,[177,161,172,310,163,161,163,163,161],[177,165,172,350,163,161,163,163,161],0.242,[182,161,161,352,163,161,163,163,161],0.003,[182,165,161,354,163,161,163,163,161],0.005,[182,161,165,341,163,161,163,163,161],[182,165,165,357,163,161,163,163,161],0.061,[182,161,172,359,163,161,163,163,161],0.111,[182,165,172,361,163,161,163,163,161],0.414,[187,161,161,352,163,161,163,163,161],[187,165,161,354,163,161,163,163,161],[187,161,165,365,163,161,163,163,161],0.008,[187,165,165,367,163,161,163,163,161],0.022,[187,161,172,318,163,161,163,163,161],[187,165,172,370,163,161,163,163,161],0.274,[192,161,161,352,163,161,163,163,161],[192,165,161,354,163,161,163,163,161],[192,161,165,374,163,161,163,163,161],0.006,[192,165,165,376,163,161,163,163,161],0.011,[192,161,172,365,163,161,163,163,161],[192,165,172,337,163,161,163,163,161],[],[105],[],[],[384],"25 synthetic range-image pairs per noise level; RMSE of ground-truth correspondence distances, unit surface diameter; GoICP variants on 1,000 points",{"slug":386,"group":387,"sourceId":271,"sourceLabel":272,"table":388,"selfRows":221,"metrics":389,"seqs":392,"entrants":406,"cells":414,"outcomes":493,"locators":494,"hardware":495,"wordings":497,"notes":498},"zhou2016fgr-table-2","zhou2016fgr:Table 2","Table 2",[390],{"label":391,"unit":110,"statistic":111,"alignment":38},"running time per pairwise registration (s)",[393,396,398,400,402,404],{"dataset":394,"sequence":395,"environment":283},"Synthetic range images","Bimba (9,416 points avg)",{"dataset":394,"sequence":397,"environment":283},"Children (11,148 points avg)",{"dataset":394,"sequence":399,"environment":283},"Dragon (11,232 points avg)",{"dataset":394,"sequence":401,"environment":283},"Angel (12,072 points avg)",{"dataset":394,"sequence":403,"environment":283},"Bunny (13,357 points avg)",{"dataset":394,"sequence":405,"environment":283},"Average (11,445 points avg)",[407,408,409,410,411,412,413],{"name":290,"methodId":5,"linkable":152,"proposed":73,"self":152},{"name":292,"methodId":5,"linkable":152,"proposed":73,"self":152},{"name":296,"methodId":82,"linkable":73,"proposed":73,"self":73},{"name":294,"methodId":82,"linkable":73,"proposed":73,"self":73},{"name":298,"methodId":299,"linkable":152,"proposed":73,"self":73},{"name":301,"methodId":82,"linkable":73,"proposed":73,"self":73},{"name":303,"methodId":271,"linkable":152,"proposed":152,"self":73},[415,417,419,421,423,425,427,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,457,459,461,463,465,466,468,469,470,472,474,476,478,480,482,484,486,488,489,491],[161,161,161,416,163,161,161,163,161],19.3,[165,161,161,418,163,161,161,163,161],19.4,[172,161,161,420,163,161,161,163,161],41,[177,161,161,422,163,161,161,163,161],311.4,[182,161,161,424,163,161,161,163,161],18.2,[187,161,161,426,163,161,161,163,161],12.8,[192,161,161,308,163,161,161,163,161],[161,161,165,429,163,161,161,163,161],21,[165,161,165,431,163,161,161,163,161],19.2,[172,161,165,433,163,161,161,163,161],136.3,[177,161,165,435,163,161,161,163,161],238.2,[182,161,165,437,163,161,161,163,161],4.8,[187,161,165,439,163,161,161,163,161],6.6,[192,161,165,441,163,161,161,163,161],0.2,[161,161,172,443,163,161,161,163,161],94.1,[165,161,172,445,163,161,161,163,161],38.4,[172,161,172,447,163,161,161,163,161],57.7,[177,161,172,449,163,161,161,163,161],483.7,[182,161,172,451,163,161,161,163,161],8.6,[187,161,172,453,163,161,161,163,161],11.9,[192,161,172,455,163,161,161,163,161],0.23,[161,161,177,429,163,161,161,163,161],[165,161,177,458,163,161,161,163,161],20.4,[172,161,177,460,163,161,161,163,161],80.9,[177,161,177,462,163,161,161,163,161],171.5,[182,161,177,464,163,161,161,163,161],8.7,[187,161,177,185,163,161,161,163,161],[192,161,177,467,163,161,161,163,161],0.26,[161,161,182,224,163,161,161,163,161],[165,161,182,180,163,161,161,163,161],[172,161,182,471,163,161,161,163,161],12.3,[177,161,182,473,163,161,161,163,161],283.8,[182,161,182,475,163,161,161,163,161],55.6,[187,161,182,477,163,161,161,163,161],12.7,[192,161,182,479,163,161,161,163,161],0.28,[161,161,187,481,163,161,161,163,161],46,[165,161,187,483,163,161,161,163,161],34,[172,161,187,485,163,161,161,163,161],65.6,[177,161,187,487,163,161,161,163,161],297.7,[182,161,187,431,163,161,161,163,161],[187,161,187,490,163,161,161,163,161],11.1,[192,161,187,492,163,161,161,163,161],0.22,[],[388],[496],"Intel Core i7-5960X 3.00 GHz, single thread",[],[499],"Average running time of each global method on each synthetic model and over all models (the number of tests averaged per model is not stated; each model has five range-image pairs at three noise levels); GoICP variants on 1,000 downsampled points, others at full resolution; single thread",{"slug":501,"group":502,"sourceId":5,"sourceLabel":6,"table":503,"selfRows":197,"metrics":504,"seqs":517,"entrants":526,"cells":531,"outcomes":545,"locators":547,"hardware":549,"wordings":550,"notes":551},"yang2016goicp-text-sec-6-2","yang2016goicp:Text Sec. 6.2","Text Sec. 6.2",[505,508,511,514,516],{"label":506,"unit":507,"statistic":90,"alignment":38},"correct registration","%",{"label":509,"unit":510,"statistic":113,"alignment":38},"all rotation errors less than","deg",{"label":512,"unit":513,"statistic":113,"alignment":38},"all translation errors less than","normalized units (points in [-1, 1]^3)",{"label":515,"unit":110,"statistic":111,"alignment":38},"mean\u002Flongest running time",{"label":515,"unit":110,"statistic":113,"alignment":38},[518,521,524],{"dataset":519,"sequence":520,"environment":118},"Stanford bunny and dragon","all 2,000 tasks",{"dataset":522,"sequence":523,"environment":118},"Stanford bunny","1,000 tests",{"dataset":525,"sequence":523,"environment":118},"Stanford dragon",[527,529],{"name":528,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT and kd-tree)",{"name":530,"methodId":5,"linkable":152,"proposed":152,"self":152},"Go-ICP (DT)",[532,534,535,537,539,541,543],[161,161,161,533,163,161,163,163,161],100,[161,165,161,172,161,161,163,163,165],[161,172,161,536,161,161,163,163,165],0.01,[165,177,165,538,163,161,161,163,172],1.6,[165,182,165,540,163,161,161,163,172],22.3,[165,177,172,542,163,161,161,163,172],1.5,[165,182,172,544,163,161,161,163,172],28.9,[546],"upper bound",[548],"Sec. 6.2",[263],[],[552,553,554],"Partial scans registered to full reconstructed models: 10 bunny scans and 10 dragon scans, 100 random initial poses each (2,000 tasks), N = 1000 data points, epsilon = 0.001 x N, domain [-pi, pi]^3 x [-0.5, 0.5]^3","Same setting as other Text Sec. 6.2 rows","Same setting as other Text Sec. 6.2 rows; 1,000 data points vs 20,000 to 40,000 model points",[556,562,568,574,580],{"group":557,"slug":558,"sourceLabel":559,"table":105,"selfRows":192,"datasets":560},"zhang2024globalbimreg:Table 1","zhang2024globalbimreg-table-1","Zhang et al., 2024b",[561],"ISPRS benchmark on indoor modelling",{"group":563,"slug":564,"sourceLabel":6,"table":565,"selfRows":182,"datasets":566},"yang2016goicp:Text Sec. 6.4","yang2016goicp-text-sec-6-4","Text Sec. 6.4",[567],"camera localization dataset [68], office scene",{"group":569,"slug":570,"sourceLabel":6,"table":571,"selfRows":177,"datasets":572},"yang2016goicp:Text Sec. 6.3","yang2016goicp-text-sec-6-3","Text Sec. 6.3",[573],"10 point-set pairs of Table 1",{"group":575,"slug":576,"sourceLabel":272,"table":577,"selfRows":172,"datasets":578},"zhou2016fgr:Table 4","zhou2016fgr-table-4","Table 4",[579],"UWA benchmark",{"group":581,"slug":582,"sourceLabel":583,"table":584,"selfRows":165,"datasets":585},"yang2021teaser:Text Sec. XI-C","yang2021teaser-text-sec-xi-c","Yang et al., 2021","Text Sec. XI-C",[586],"Stanford Bunny (100 points)",1790510665037]