[{"data":1,"prerenderedAt":649},["ShallowReactive",2],{"method-kintinuous2015":3},{"method":4,"reference":66,"equipment":90,"figures":120,"results":121},{"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":23,"limitations":29,"sensors":36,"platform":38,"estimator":40,"association":41,"timeModel":42,"deskew":43,"loopClosure":44,"globalOptimization":45,"mapRepresentation":46,"prior":47,"outputGeometry":48,"compute":49,"codeUrl":50,"codeLicense":51,"relatedVersions":52},"kintinuous2015","Whelan et al., 2015b","Kintinuous","Real-time large-scale dense RGB-D SLAM with volumetric fusion",2015,"classic","C08","full_slam_with_global_correction","Kintinuous 以 GPU 上的循環緩衝（cyclical buffer）讓 TSDF 融合體積隨相機移動，使稠密融合可延伸到無界空間，並結合稠密幾何與光度約束估計位姿。偵測到迴圈後，以 as-rigid-as-possible 空間變形校正已建立的稠密地圖。作者報告可在數百公尺範圍內即時產生全域一致的表面重建。","Kintinuous shifts a TSDF volume with the camera via a GPU cyclical buffer, combines dense geometric and photometric tracking, and corrects the dense map after loop closure with as-rigid-as-possible deformation.","full_text_reviewed","peer_reviewed_published","background","論文未報告營建現場測試；資料為作者自錄室內外序列與 RGB-D 基準。重複經過區域不重新融合而產生重疊（aliasing）的限制，對需要多次行經的工地掃描有直接意義（推論）。",[20,21,22],"public_benchmark","controlled_experiment","simulation",[24,25,26,27,28],"Globally consistent surface reconstructions over hundreds of metres with a commodity RGB-D sensor (abstract)","Fused volumetric map gives higher quality than raw RGB-D point clouds (abstract)","Deformation-corrected maps agree with 2-pass reconstructions to 1.2 to 2.8 mm residual on six datasets and 19.0 mm on Apartment (Table 5)","No post-processing is needed, whereas DVO SLAM needed 7.41 to 5501.58 s of post-processing and failed on five of the seven hand-held datasets (Table 6)","ATE RMSE on TUM RGB-D typically within 3 cm of DVO SLAM, RGB-D SLAM and MRS (Sec. 5.1.1, Table 2)",[30,31,32,33,34,35],"Projective data association limits the motions the odometry front-end can handle (conclusion)","Revisited areas are not re-integrated into the volume, causing aliasing where multiple passes occur (conclusion)","Higher ATE on fr1\u002Fdesk2 and fr1\u002Froom, linked to high angular velocity, motion blur and rolling shutter (Sec. 5.1, Table 1)","Surface self-intersection is possible after deformation, though rarely observed (Sec. 5.2)","Fixed-step raycasting causes artefacts near object edges that lower synthetic depth accuracy (Sec. 5.2.3)","Volume placement can leave distant surfaces such as a ceiling unreconstructed (Fig. 17 caption)",[37],"RGB-D",[39],"handheld","Frame-to-model point-to-plane ICP against the raycast TSDF combined with frame-to-frame dense photometric RGB-D alignment in a weighted sum (w_rgbd = 0.1), three-level pyramids, GPU tree reduction and CPU Cholesky solve; loop constraints enter an iSAM pose graph, whose optimised poses and matched SURF points constrain an embedded-deformation optimisation (weights 1, 10, 100, 100) solved by Gauss-Newton with CHOLMOD","dense geometric and photometric; projective data association (conclusion)","discrete poses","not_reported (rolling shutter not modelled; authors note that projective data association limits the camera motions the front-end can handle, which also limits motion blur and rolling-shutter effects, and that real-time correction would add computation; conclusion)","Frames enter the DBoW (SURF) database when a combined rotation and translation motion metric exceeds 0.3; a candidate needs at least 35 FLANN SURF matches, a RANSAC 3-point transform with a 2.0 px reprojection threshold and at least 25% inliers refined by Levenberg-Marquardt, and a final ICP between voxel-downsampled clouds accepted when the mean squared correspondence error is below 0.01; the accepted constraint is added to iSAM and the dense map is corrected by as-rigid-as-possible embedded deformation","iSAM incremental pose-graph optimisation plus non-rigid embedded deformation of all cloud-slice vertices; deformation nodes are sampled along the pose graph with sequential k = 4 connectivity and vertices are associated by back-traversal so that unrelated map regions are not linked; runs online without a final batch step, optionally on a subsampled pose graph","GPU TSDF of 512^3 voxels (6 bytes each: truncated float16 distance, uint8 weight, RGB) addressed with modulo arithmetic as a cyclical buffer that shifts with the camera; surface leaving the volume is extracted by axis-aligned raycasts into voxel-grid-filtered cloud slices tied to the pose that caused the shift and incrementally triangulated with Greedy Projection Triangulation; revisited areas are not re-fused","none","Large-scale dense coloured surface as cloud slices and a triangle mesh; the seven hand-held datasets span 30 to 318 m and about 0.9 to 6.2 million vertices","Desktop PC (Ubuntu 12.04) with Intel Core i7-3960X at 3.30 GHz, 16 GB RAM and nVidia GeForce 680GTX with 2 GB; the 512^3-voxel TSDF uses 768 MB of GPU memory. The frontend averages 29.94 ms per frame on fr1\u002Fdesk at the chosen shift threshold of 16 voxels, below the 30 Hz sensor period; loop-closure latency (recognition to corrected map) is 0.99 to 11.56 s with an every-frame pose graph and 0.64 to 2.78 s with a subsampled pose graph over the six datasets","https:\u002F\u002Fgithub.com\u002Fmp3guy\u002FKintinuous","custom licence, non-commercial purposes only (LICENSE.txt)",[53,55,59,62],{"relation":54,"title":7,"doi_or_url":50},"code_release",{"relation":56,"title":57,"doi_or_url":58},"conference_version","Robust real-time visual odometry for dense RGB-D mapping (ICRA 2013, pp. 5724-5731)","10.1109\u002FICRA.2013.6631400",{"relation":56,"title":60,"doi_or_url":61},"Deformation-based loop closure for large scale dense RGB-D SLAM (IROS 2013, pp. 548-555)","10.1109\u002FIROS.2013.6696405",{"relation":63,"title":64,"doi_or_url":65},"workshop_version","Kintinuous: Spatially Extended KinectFusion (RSS RGB-D Workshop 2012; no DOI)","http:\u002F\u002Fthomaswhelan.ie\u002FWhelan12rssw.pdf",{"id":5,"kind":67,"shortName":7,"title":8,"authors":68,"year":9,"venue":75,"venueType":76,"publisher":77,"volumeIssuePages":78,"doi":79,"arxivId":80,"url":81,"firstPublicDate":82,"publicationStatus":16,"metadataStatus":83,"fulltextStatus":15,"era":10,"classicReason":84,"codeUrl":50,"cluster":11,"topics":85,"mdpi":86,"verification":87,"label":6,"fulltextRoute":88,"versionRead":89,"addedByCensus":86},"method",[69,70,71,72,73,74],"Thomas Whelan","Michael Kaess","Hordur Johannsson","Maurice Fallon","John J. Leonard","John McDonald","The International Journal of Robotics Research","journal","SAGE","34(4-5):598-626","10.1177\u002F0278364914551008",null,"https:\u002F\u002Fwww.thomaswhelan.ie\u002FWhelan14ijrr.pdf","2014-12-09","metadata_verified","necessary technical node: extends TSDF fusion to unbounded space (cyclical GPU buffer) and corrects dense maps after loop closure via as-rigid-as-possible deformation.",[11],false,"corrected","author copy","Author's final manuscript (29 PDF pages) from the first author's website; byte-identical to the MIT DSpace deposit hdl:1721.1\u002F97583 (Leonard_Real-time.pdf, 32,264,016 bytes, MD5 4a07f7b3679bcfc4ee9cd73544bccb38 in the DSpace REST metadata equals the MD5 of the downloaded file), labelled Author's final manuscript under CC BY-NC-SA 4.0; SAGE version of record not compared",[91,97,101,108,113],{"category":92,"model":93,"canonical":93,"role":94,"dataset":80,"specs":95,"locator":96},"compute","Intel Core i7-3960X","compute for runtime","3.30 GHz, 16 GB RAM, Ubuntu 12.04 desktop","Sec. 5.3",{"category":92,"model":98,"canonical":98,"role":94,"dataset":80,"specs":99,"locator":100},"nVidia GeForce 680GTX","2 GB GPU memory; the 512^3 TSDF volume uses 768 MB","Sec. 2.2, 5.3",{"category":102,"model":103,"canonical":103,"role":104,"dataset":105,"specs":106,"locator":107},"rgbd","commodity RGB-D camera (model not stated)","method input","authors' seven hand-held datasets","640x480 frames at 30 Hz; auto exposure and auto white balance enabled in all real datasets","Sec. 2.5.1, 5.1, 5.3.1",{"category":109,"model":110,"canonical":110,"role":104,"dataset":105,"specs":111,"locator":112},"platform","hand-held RGB-D camera","seven datasets from 30 to 318 m (coffee room, corridor, garden, outdoors, two floors, indoor and outdoor, apartment)","Sec. 5.2; Table 5",{"category":114,"model":115,"canonical":115,"role":116,"dataset":117,"specs":118,"locator":119},"other","motion capture system (TUM RGB-D ground truth, model not stated)","reference or ground truth","TUM RGB-D","synchronised ground-truth poses","Sec. 5.1",[],{"totalRows":122,"groupCount":123,"groups":124,"others":459},264,41,[125,254,332,386],{"slug":126,"group":127,"sourceId":5,"sourceLabel":6,"table":128,"selfRows":129,"metrics":130,"seqs":142,"entrants":173,"cells":176,"outcomes":248,"locators":249,"hardware":250,"wordings":251,"notes":252},"kintinuous2015-table-1","kintinuous2015:Table 1","Table 1",30,[131,136,139],{"label":132,"unit":133,"statistic":134,"alignment":135},"ATE RMSE (m)","m","RMSE","not_reported",{"label":137,"unit":133,"statistic":138,"alignment":135},"ATE Median (m)","median",{"label":140,"unit":133,"statistic":141,"alignment":135},"ATE Max (m)","max",[143,146,149,152,155,158,161,164,167,170],{"dataset":117,"sequence":144,"environment":145},"fr1\u002Fdesk","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 23.33 deg\u002Fs",{"dataset":117,"sequence":147,"environment":148},"fr1\u002Fdesk2","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 29.31 deg\u002Fs",{"dataset":117,"sequence":150,"environment":151},"fr1\u002Froom","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 29.88 deg\u002Fs",{"dataset":117,"sequence":153,"environment":154},"fr1\u002Fxyz","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 8.92 deg\u002Fs",{"dataset":117,"sequence":156,"environment":157},"fr1\u002Frpy","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 50.15 deg\u002Fs",{"dataset":117,"sequence":159,"environment":160},"fr1\u002Fplant","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 27.89 deg\u002Fs",{"dataset":117,"sequence":162,"environment":163},"fr2\u002Fdesk","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 6.34 deg\u002Fs",{"dataset":117,"sequence":165,"environment":166},"fr2\u002Fxyz","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 1.72 deg\u002Fs",{"dataset":117,"sequence":168,"environment":169},"fr3\u002Flong","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 10.19 deg\u002Fs",{"dataset":117,"sequence":171,"environment":172},"fr3\u002Fnst","TUM RGB-D benchmark sequences (RGB-D sensor moved through an environment, motion-capture ground truth; Sec. 5.1); mean angular velocity 7.43 deg\u002Fs",[174],{"name":7,"methodId":5,"linkable":175,"proposed":175,"self":175},true,[177,181,184,187,189,191,193,195,197,199,202,204,206,209,211,213,216,218,220,223,225,227,230,232,234,237,239,241,244,246],[178,178,178,179,180,178,180,180,178],0,0.0407,-1,[178,182,178,183,180,178,180,180,178],1,0.0352,[178,185,178,186,180,178,180,180,178],2,0.0905,[178,178,182,188,180,178,180,180,178],0.0747,[178,182,182,190,180,178,180,180,178],0.0639,[178,185,182,192,180,178,180,180,178],0.2309,[178,178,185,194,180,178,180,180,178],0.0813,[178,182,185,196,180,178,180,180,178],0.0739,[178,185,185,198,180,178,180,180,178],0.2511,[178,178,200,201,180,178,180,180,178],3,0.018,[178,182,200,203,180,178,180,180,178],0.0155,[178,185,200,205,180,178,180,180,178],0.0392,[178,178,207,208,180,178,180,180,178],4,0.0311,[178,182,207,210,180,178,180,180,178],0.0213,[178,185,207,212,180,178,180,180,178],0.0991,[178,178,214,215,180,178,180,180,178],5,0.05,[178,182,214,217,180,178,180,180,178],0.0425,[178,185,214,219,180,178,180,180,178],0.1148,[178,178,221,222,180,178,180,180,178],6,0.0376,[178,182,221,224,180,178,180,180,178],0.0315,[178,185,221,226,180,178,180,180,178],0.0879,[178,178,228,229,180,178,180,180,178],7,0.0341,[178,182,228,231,180,178,180,180,178],0.0234,[178,185,228,233,180,178,180,180,178],0.0979,[178,178,235,236,180,178,180,180,178],8,0.0329,[178,182,235,238,180,178,180,180,178],0.0297,[178,185,235,240,180,178,180,180,178],0.0698,[178,178,242,243,180,178,180,180,178],9,0.0372,[178,182,242,245,180,178,180,180,178],0.0335,[178,185,242,247,180,178,180,180,178],0.0735,[],[128],[],[],[253],"TUM RGB-D ATE statistics of Kintinuous (m), mean over ten runs of each dataset; mean angular velocity (deg\u002Fs) of each sequence given in the table",{"slug":255,"group":256,"sourceId":257,"sourceLabel":258,"table":259,"selfRows":260,"metrics":261,"seqs":270,"entrants":281,"cells":286,"outcomes":326,"locators":327,"hardware":328,"wordings":329,"notes":330},"handa2014iclnuim-table-ii","handa2014iclnuim:Table II","handa2014iclnuim","Handa et al., 2014","Table II",16,[262,266,267,269],{"label":263,"unit":133,"statistic":264,"alignment":265},"cloud\u002Fmesh distance: perpendicular distance from each reconstructed vertex to the closest ground-truth model triangle (CloudCompare)","mean","SE3",{"label":263,"unit":133,"statistic":138,"alignment":265},{"label":263,"unit":133,"statistic":268,"alignment":265},"std",{"label":263,"unit":133,"statistic":141,"alignment":265},[271,275,277,279],{"dataset":272,"sequence":273,"environment":274},"ICL-NUIM","kt0 (lr)","synthetic living room (POV-Ray ray-traced)",{"dataset":272,"sequence":276,"environment":274},"kt1 (lr)",{"dataset":272,"sequence":278,"environment":274},"kt2 (lr)",{"dataset":272,"sequence":280,"environment":274},"kt3 (lr)",[282,284],{"name":283,"methodId":80,"linkable":86,"proposed":86,"self":86},"Kintinuous pipeline with DVO odometry [13]",{"name":285,"methodId":5,"linkable":175,"proposed":86,"self":175},"Kintinuous pipeline with ICP odometry (as in KinectFusion and Kintinuous [3], [4])",[287,289,291,293,295,297,299,301,303,305,307,309,311,313,315,316,318,320,322,324],[178,178,178,288,180,178,180,180,178],0.0662,[178,182,178,290,180,178,180,180,178],0.0593,[178,185,178,292,180,178,180,180,178],0.0504,[178,200,178,294,180,178,180,180,178],0.3655,[182,178,178,296,180,178,180,180,178],0.0612,[182,182,178,298,180,178,180,180,178],0.0368,[182,185,178,300,180,178,180,180,178],0.0821,[182,200,178,302,180,178,180,180,178],0.5456,[182,178,182,304,180,178,180,180,178],0.0034,[182,182,182,306,180,178,180,180,178],0.0026,[182,185,182,308,180,178,180,180,178],0.0033,[182,200,182,310,180,178,180,180,178],0.0461,[182,178,185,312,180,178,180,180,178],0.0037,[182,182,185,314,180,178,180,180,178],0.003,[182,185,185,304,180,178,180,180,178],[182,200,185,317,180,178,180,180,178],0.0508,[182,178,200,319,180,178,180,180,178],0.0085,[182,182,200,321,180,178,180,180,178],0.0063,[182,185,200,323,180,178,180,180,178],0.0072,[182,200,200,325,180,178,180,180,178],0.1562,[],[259],[],[],[331],"Surface reconstruction error, noise-free living room: CloudCompare cloud\u002Fmesh distance after manual coarse alignment and ICP fine alignment to the densely sampled model; ICP odometry except the kt0 (DVO) column; TSDF volume 4.5 m, truncation 0.045 m",{"slug":333,"group":334,"sourceId":257,"sourceLabel":258,"table":335,"selfRows":260,"metrics":336,"seqs":341,"entrants":346,"cells":348,"outcomes":380,"locators":381,"hardware":382,"wordings":383,"notes":384},"handa2014iclnuim-table-vii","handa2014iclnuim:Table VII","Table VII",[337,338,339,340],{"label":263,"unit":133,"statistic":264,"alignment":265},{"label":263,"unit":133,"statistic":138,"alignment":265},{"label":263,"unit":133,"statistic":268,"alignment":265},{"label":263,"unit":133,"statistic":141,"alignment":265},[342,343,344,345],{"dataset":272,"sequence":273,"environment":274},{"dataset":272,"sequence":276,"environment":274},{"dataset":272,"sequence":278,"environment":274},{"dataset":272,"sequence":280,"environment":274},[347],{"name":285,"methodId":5,"linkable":175,"proposed":86,"self":175},[349,351,353,355,357,359,361,363,365,366,368,370,372,374,376,378],[178,178,178,350,180,178,180,180,178],0.0114,[178,182,178,352,180,178,180,180,178],0.0084,[178,185,178,354,180,178,180,180,178],0.0171,[178,200,178,356,180,178,180,180,178],1.0377,[178,178,182,358,180,178,180,180,178],0.008,[178,182,182,360,180,178,180,180,178],0.0048,[178,185,182,362,180,178,180,180,178],0.0286,[178,200,182,364,180,178,180,180,178],1.0911,[178,178,185,319,180,178,180,180,178],[178,182,185,367,180,178,180,180,178],0.0071,[178,185,185,369,180,178,180,180,178],0.0136,[178,200,185,371,180,178,180,180,178],1.0798,[178,178,200,373,180,178,180,180,178],0.1503,[178,182,200,375,180,178,180,180,178],0.0124,[178,185,200,377,180,178,180,180,178],0.2745,[178,200,200,379,180,178,180,180,178],1.0499,[],[335],[],[],[385],"Surface reconstruction error, living room with simulated noise, all using ICP odometry; same CloudCompare cloud\u002Fmesh procedure",{"slug":387,"group":388,"sourceId":5,"sourceLabel":6,"table":389,"selfRows":390,"metrics":391,"seqs":398,"entrants":421,"cells":426,"outcomes":453,"locators":454,"hardware":455,"wordings":456,"notes":457},"kintinuous2015-table-5","kintinuous2015:Table 5","Table 5",14,[392,396],{"label":393,"unit":394,"statistic":134,"alignment":395},"2-pass residual registration error (mm)","mm","not_applicable",{"label":397,"unit":394,"statistic":134,"alignment":395},"2-pass fast residual registration error (mm)",[399,403,406,409,412,415,418],{"dataset":400,"sequence":401,"environment":402},"authors' hand-held datasets","Coffee (30.18 m, 909422 vertices)","small coffee room",{"dataset":400,"sequence":404,"environment":405},"Indoors (49.57 m, 1603116 vertices)","corridor loop",{"dataset":400,"sequence":407,"environment":408},"Garden (71.49 m, 2418331 vertices)","large cluttered outdoor area",{"dataset":400,"sequence":410,"environment":411},"Outdoors (152.05 m, 2961966 vertices)","large outdoor area with brickwork",{"dataset":400,"sequence":413,"environment":414},"Two floors (173.88 m, 4016273 vertices)","two floors of a building",{"dataset":400,"sequence":416,"environment":417},"In\u002Foutdoors (317.95 m, 5985669 vertices)","indoor and outdoor",{"dataset":400,"sequence":419,"environment":420},"Apartment (61.27 m, 6205222 vertices)","two floors of an apartment",[422,424],{"name":423,"methodId":5,"linkable":175,"proposed":175,"self":175},"Kintinuous (every-frame pose graph)",{"name":425,"methodId":5,"linkable":175,"proposed":175,"self":175},"Kintinuous (subsampled pose graph)",[427,429,431,433,435,437,439,441,443,444,445,447,449,451],[178,178,178,428,180,178,180,180,178],1.2,[182,182,178,430,180,178,180,180,178],1.8,[178,178,182,432,180,178,180,180,178],2.7,[182,182,182,434,180,178,180,180,178],4.9,[178,178,185,436,180,178,180,180,178],2.1,[182,182,185,438,180,178,180,180,178],2.5,[178,178,200,440,180,178,180,180,178],2.3,[182,182,200,442,180,178,180,180,178],8.8,[178,178,207,436,180,178,180,180,178],[182,182,207,235,180,178,180,180,178],[178,178,214,446,180,178,180,180,178],2.8,[182,182,214,448,180,178,180,180,178],7.5,[178,178,221,450,180,178,180,180,178],19,[182,182,221,452,180,178,180,180,178],20.4,[],[389],[],[],[458],"Seven hand-held datasets: RMS residual of point-to-plane ICP between the deformation-corrected map and a 2-pass map rebuilt from the optimised pose graph (mm); '2-pass fast' uses a subsampled pose graph; a consistency measure, not accuracy against an independent reference",[460,466,473,478,483,488,493,497,502,507,512,518,524,532,537,542,547,552,557,562,568,573,577,583,587,591,596,600,604,609,614,618,622,627,632,637,642],{"group":461,"slug":462,"sourceLabel":6,"table":463,"selfRows":464,"datasets":465},"kintinuous2015:Table 2","kintinuous2015-table-2","Table 2",10,[117],{"group":467,"slug":468,"sourceLabel":469,"table":470,"selfRows":235,"datasets":471},"choi2015robustrecon:Table 7","choi2015robustrecon-table-7","Choi et al., 2015","Table 7",[472],"SUN3D",{"group":474,"slug":475,"sourceLabel":258,"table":476,"selfRows":235,"datasets":477},"handa2014iclnuim:Table III","handa2014iclnuim-table-iii","Table III",[272],{"group":479,"slug":480,"sourceLabel":258,"table":481,"selfRows":235,"datasets":482},"handa2014iclnuim:Table IV","handa2014iclnuim-table-iv","Table IV",[272],{"group":484,"slug":485,"sourceLabel":258,"table":486,"selfRows":235,"datasets":487},"handa2014iclnuim:Table V","handa2014iclnuim-table-v","Table V",[272],{"group":489,"slug":490,"sourceLabel":258,"table":491,"selfRows":235,"datasets":492},"handa2014iclnuim:Table VI","handa2014iclnuim-table-vi","Table VI",[272],{"group":494,"slug":495,"sourceLabel":6,"table":470,"selfRows":228,"datasets":496},"kintinuous2015:Table 7","kintinuous2015-table-7",[117],{"group":498,"slug":499,"sourceLabel":6,"table":500,"selfRows":228,"datasets":501},"kintinuous2015:Table 8","kintinuous2015-table-8","Table 8",[400],{"group":503,"slug":504,"sourceLabel":6,"table":505,"selfRows":228,"datasets":506},"kintinuous2015:Table 9","kintinuous2015-table-9","Table 9",[400],{"group":508,"slug":509,"sourceLabel":510,"table":476,"selfRows":228,"datasets":511},"orbslam2_2017:Table III","orbslam2-2017-table-iii","Mur-Artal & Tardos, 2017",[117],{"group":513,"slug":514,"sourceLabel":515,"table":463,"selfRows":221,"datasets":516},"loopyslam2024:Table 2","loopyslam2024-table-2","Liso et al., 2024",[517],"TUM-RGBD",{"group":519,"slug":520,"sourceLabel":521,"table":522,"selfRows":221,"datasets":523},"pointslam2023:Table 3","pointslam2023-table-3","Sandström et al., 2023","Table 3",[517],{"group":525,"slug":526,"sourceLabel":527,"table":128,"selfRows":221,"datasets":528},"splatam2024:Table 1","splatam2024-table-1","Keetha et al., 2024",[529,530,531,517],"Replica","ScanNet (original)","ScanNet++",{"group":533,"slug":534,"sourceLabel":6,"table":535,"selfRows":214,"datasets":536},"kintinuous2015:Table 10","kintinuous2015-table-10","Table 10",[400],{"group":538,"slug":539,"sourceLabel":6,"table":540,"selfRows":214,"datasets":541},"kintinuous2015:Table 11","kintinuous2015-table-11","Table 11",[400],{"group":543,"slug":544,"sourceLabel":545,"table":522,"selfRows":207,"datasets":546},"bundlefusion2017:Table 3","bundlefusion2017-table-3","Dai et al., 2017a",[272],{"group":548,"slug":549,"sourceLabel":545,"table":550,"selfRows":207,"datasets":551},"bundlefusion2017:Table 4","bundlefusion2017-table-4","Table 4",[117],{"group":553,"slug":554,"sourceLabel":545,"table":555,"selfRows":207,"datasets":556},"bundlefusion2017:Table 6","bundlefusion2017-table-6","Table 6",[272],{"group":558,"slug":559,"sourceLabel":545,"table":470,"selfRows":207,"datasets":560},"bundlefusion2017:Table 7","bundlefusion2017-table-7",[561],"Augmented ICL-NUIM",{"group":563,"slug":564,"sourceLabel":469,"table":565,"selfRows":207,"datasets":566},"choi2015robustrecon:Supp. Table 3","choi2015robustrecon-supp-table-3","Supp. Table 3",[567],"augmented ICL-NUIM (synthetic, realistic noise, full-scan trajectories)",{"group":569,"slug":570,"sourceLabel":469,"table":571,"selfRows":207,"datasets":572},"choi2015robustrecon:Supp. Table 4","choi2015robustrecon-supp-table-4","Supp. Table 4",[567],{"group":574,"slug":575,"sourceLabel":469,"table":550,"selfRows":207,"datasets":576},"choi2015robustrecon:Table 4","choi2015robustrecon-table-4",[567],{"group":578,"slug":579,"sourceLabel":580,"table":581,"selfRows":207,"datasets":582},"elasticfusion2015:Table I","elasticfusion2015-table-i","Whelan et al., 2015a","Table I",[117],{"group":584,"slug":585,"sourceLabel":580,"table":259,"selfRows":207,"datasets":586},"elasticfusion2015:Table II","elasticfusion2015-table-ii",[272],{"group":588,"slug":589,"sourceLabel":580,"table":476,"selfRows":207,"datasets":590},"elasticfusion2015:Table III","elasticfusion2015-table-iii",[272],{"group":592,"slug":593,"sourceLabel":594,"table":463,"selfRows":207,"datasets":595},"gsslam2024:Table 2","gsslam2024-table-2","Yan et al., 2024",[117],{"group":597,"slug":598,"sourceLabel":6,"table":522,"selfRows":207,"datasets":599},"kintinuous2015:Table 3","kintinuous2015-table-3",[117],{"group":601,"slug":602,"sourceLabel":6,"table":550,"selfRows":207,"datasets":603},"kintinuous2015:Table 4","kintinuous2015-table-4",[117],{"group":605,"slug":606,"sourceLabel":607,"table":128,"selfRows":207,"datasets":608},"monogs2024:Table 1","monogs2024-table-1","Matsuki et al., 2024",[117],{"group":610,"slug":611,"sourceLabel":612,"table":463,"selfRows":207,"datasets":613},"psmslam2017:Table 2","psmslam2017-table-2","Yan et al., 2017",[117],{"group":615,"slug":616,"sourceLabel":612,"table":550,"selfRows":207,"datasets":617},"psmslam2017:Table 4","psmslam2017-table-4",[272],{"group":619,"slug":620,"sourceLabel":612,"table":389,"selfRows":207,"datasets":621},"psmslam2017:Table 5","psmslam2017-table-5",[272],{"group":623,"slug":624,"sourceLabel":625,"table":463,"selfRows":200,"datasets":626},"badslam2019:Table 2","badslam2019-table-2","Schöps et al., 2019",[117],{"group":628,"slug":629,"sourceLabel":630,"table":550,"selfRows":200,"datasets":631},"coslam2023:Table 4","coslam2023-table-4","Wang et al., 2023a",[117],{"group":633,"slug":634,"sourceLabel":635,"table":522,"selfRows":200,"datasets":636},"imap2021:Table 3","imap2021-table-3","Sucar et al., 2021",[117],{"group":638,"slug":639,"sourceLabel":640,"table":463,"selfRows":200,"datasets":641},"niceslam2022:Table 2","niceslam2022-table-2","Zhu et al., 2022a",[117],{"group":643,"slug":644,"sourceLabel":645,"table":646,"selfRows":185,"datasets":647},"droidslam2021:Fig. 4 table","droidslam2021-fig-4-table","Teed & Deng, 2021","Fig. 4 table",[648],"ETH3D SLAM benchmark",1790510658354]