[{"data":1,"prerenderedAt":679},["ShallowReactive",2],{"method-mrsmap2014":3},{"method":4,"reference":60,"equipment":81,"figures":103,"results":104},{"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":35,"platform":37,"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},"mrsmap2014","Stückler & Behnke, 2014","MRSMap","Multi-resolution surfel maps for efficient dense 3D modeling and tracking",2014,"classic","C08","full_slam_with_global_correction","MRSMap 把每張 RGB-D 影像轉成八元樹多解析度面元地圖：各層節點都以單次掃描累加的充分統計量，保存點位置與 Lαβ 色彩的六維常態分布，並依最多六個觀測方向分開保存面元；最細解析度隨深度平方放寬，以反映 RGB-D 深度雜訊。配準時由最細解析度開始，在鄰域中尋找形狀與紋理描述子相符的面元配對，先以 Levenberg-Marquardt 初始化、再以牛頓法最大化配對似然，並以閉式近似估計位姿共變異。SLAM 以關鍵視角為節點，每一影格隨機抽選一個鄰近關鍵視角嘗試配準以發現迴圈，再以 g2o 最佳化位姿圖，全程在 CPU 上即時執行，同一方法也用於物件建模與追蹤。","CPU RGB-D registration and key-view SLAM on octree multi-resolution surfel maps: each node stores a 6D position-colour Gaussian per view direction with depth-adapted finest resolution; maps are registered by descriptor-gated multi-resolution surfel association with LM then Newton optimization and closed-form pose covariance, and key views are linked by randomized loop-closure tests and g2o pose-graph optimization.","full_text_reviewed","peer_reviewed_published","background","論文未在施工現場測試，評估為 TUM (Freiburg) 室內小場景與桌上物件資料集，並在 RoboCup@Home 競賽中讓服務機器人即時追蹤桌子與鍋具。在 CPU 上以多解析度面元配準 RGB-D 影像，適合運算資源有限的室內機器人；但論文 Table 3 的 SLAM 精度已由 2015 年更正啟事（JVCIR 26:349）修訂，引用時應採更正後數值。corpus 中 bundlefusion2017、elasticfusion2015、kintinuous2015 與 badslam2019 皆以 MRSMap 作為比較基準。",[20,21,22],"public_benchmark","controlled_experiment","task_level_validation",[24,25,26,27,28],"Lowest median incremental-registration RPE on most of 22 TUM sequences against warp, GICP, 3D-NDT and fovis (Table 1)","About 15 Hz at VGA on a notebook CPU: 75 ms versus 4015 ms for GICP and 415 ms for 3D-NDT on fr1 desk (Table 2)","Keeps ICP-like robustness for larger frame skips where warp fails (Fig. 6)","Object models with median ATE of about 1 to 2 cm, and graph optimization strongly reduces error (Table 4)","Used for real-time object tracking on the Cosero robot at RoboCup@Home 2011 and 2012 (Sec. 6.4)",[30,31,32,33,34],"Did not succeed on freiburg1_floor (little texture) or freiburg2_large_loop, and loses track when much of the image has invalid or highly uncertain distant depth (Sec. 6.2)","Real-time SLAM must drop frames and limit the maximum resolution to 0.05 m (Sec. 6.2)","Loop detection relies on pose proximity; appearance-based loop detection for larger scenes is future work (Sec. 5.1.1, Sec. 7)","Fovis is much faster (13 to 16 ms) and more accurate on some textured sequences (Tables 1 and 2)","The SLAM RPE values of Table 3 were corrected by a 2015 corrigendum; with the corrected values the method beats RGB-D SLAM on 8 of 11 sequences rather than the nine stated in Sec. 6.2 (inference from the corrigendum)",[36],"RGB-D camera at VGA 640x480 and 30 Hz (TUM Freiburg benchmark sequences and the authors' object dataset; sensor model not named); QVGA used on the robot",[38,39],"RGB-D camera moved through the TUM Freiburg scenes and around objects for the authors' object dataset (carrying mode not stated, Secs. 5 and 6)","mobile manipulation robot Cosero (RoboCup@Home 2011 and 2012 demonstrations; locomotion not described, Sec. 6.4)","Key-view SLAM: each frame is registered to the current reference key view by maximizing the surfel-match likelihood (approximate Levenberg-Marquardt initialization, then Newton's method with trilinear interpolation, typically 10 to 20 LM and 5 Newton iterations); key views linked by relative-pose constraints with closed-form covariance; pose graph solved by sparse Cholesky in g2o, one iteration per frame","Multi-resolution surfel association starting at the finest resolution with a cubic volume query around the transformed surfel mean (side twice the node resolution), bootstrapped from previous associations via the 26-neighbourhood; accepted only if shape-texture descriptors (surfel-pair angle histograms and luminance and chrominance contrasts) differ by at most 0.1 and contour flags agree","discrete poses","not_applicable (RGB-D input)","Randomized hypothesis-and-test: per frame one key view, sampled with probability decreasing with distance and angle from the reference, is registered; the constraint is accepted if its bidirectional matching likelihood is at least a fraction of that of the key view's initial constraint","Pose graph over key views in g2o, iterated once per frame (median 0.79 ms, max 4.01 ms on freiburg2_desk with 64 key views and 138 edges); optimized key views fused into one multi-view map","octree multi-resolution surfel map: every node stores sufficient statistics of a 6D Gaussian of position and L-alpha-beta colour, up to six surfels per node for orthogonal view directions; finest node size adapted to squared depth (0.0125 m limit); border and occluded-background surfels excluded","none","multi-view multi-resolution surfel map of a scene or an object model (visualized by sampling the surfel distributions; object models about 54 MB for a chair and 19 MB for a humanoid)","CPU only (multi-core parallel association and derivatives); notebook Intel Core i7 3610QM 2.3 GHz; registration about 15 Hz at VGA (61 to 75 ms per frame); real-time SLAM drops frames and limits maximum resolution to 0.05 m","https:\u002F\u002Fgithub.com\u002Fjstueckler\u002Fmrsmap","BSD licence (LICENSE file checked, University of Bonn)",[53,57],{"relation":54,"title":55,"doi_or_url":56},"erratum","Corrigendum to 'Multi-resolution surfel maps for efficient dense 3D modeling and tracking', JVCIR 26 (2015) 349, which replaces Table 3","10.1016\u002Fj.jvcir.2014.10.001",{"relation":58,"title":59,"doi_or_url":50},"code_release","jstueckler\u002Fmrsmap (original release at code.google.com\u002Fp\u002Fmrsmap)",{"id":5,"kind":61,"shortName":7,"title":8,"authors":62,"year":9,"venue":65,"venueType":66,"publisher":67,"volumeIssuePages":68,"doi":69,"arxivId":70,"url":71,"firstPublicDate":72,"publicationStatus":16,"metadataStatus":73,"fulltextStatus":15,"era":10,"classicReason":74,"codeUrl":50,"cluster":11,"topics":75,"mdpi":76,"verification":77,"label":6,"fulltextRoute":78,"versionRead":79,"addedByCensus":80},"method",[63,64],"Jörg Stückler","Sven Behnke","Journal of Visual Communication and Image Representation","journal","Elsevier","25(1):137-147","10.1016\u002Fj.jvcir.2013.02.008",null,"https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1016\u002Fj.jvcir.2013.02.008","2013-03-05","metadata_verified","principle reused: RGB-D images compressed into octree multi-resolution surfel maps (per-node 6D position-colour Gaussians) that are registered on a CPU by descriptor-gated multi-resolution association with a probabilistic likelihood and pose covariance, feeding key-view pose-graph SLAM; frequent CPU surfel baseline in bundlefusion2017, elasticfusion2015, kintinuous2015 and badslam2019.",[11],false,"corrected","NTU institutional (Chrome)","ScienceDirect HTML full text of the version of record (JVCIR 25(1), 2014) plus the corrigendum (JVCIR 26, 2015, p. 349)",true,[82,88,92,97],{"category":83,"model":84,"canonical":84,"role":85,"dataset":70,"specs":86,"locator":87},"compute","notebook PC with Intel Core i7 3610QM 2.3 GHz (max. 3.3 GHz) QuadCore CPU","compute for runtime","timings of all methods at VGA resolution","Sec. 6",{"category":83,"model":89,"canonical":89,"role":85,"dataset":70,"specs":90,"locator":91},"quadcore notebook with Intel i7-Q720 CPU (Cosero's main computer)","RGB-D images subsampled to QVGA for tracking during RoboCup@Home demonstrations","Sec. 6.4",{"category":93,"model":94,"canonical":94,"role":95,"dataset":70,"specs":96,"locator":91},"platform","Cosero mobile manipulation robot","method input","used the tracking method at RoboCup@Home 2011 and 2012",{"category":98,"model":99,"canonical":99,"role":100,"dataset":101,"specs":102,"locator":87},"other","external optical motion capture systems (models not reported)","reference or ground truth","TUM RGB-D and authors' object tracking dataset","ground-truth camera poses of the TUM benchmark and the authors' object dataset",[],{"totalRows":105,"groupCount":106,"groups":107,"others":664},60,7,[108,391,504,594],{"slug":109,"group":110,"sourceId":5,"sourceLabel":6,"table":111,"selfRows":112,"metrics":113,"seqs":119,"entrants":166,"cells":180,"outcomes":385,"locators":386,"hardware":387,"wordings":388,"notes":389},"mrsmap2014-table-1","mrsmap2014:Table 1","Table 1",22,[114],{"label":115,"unit":116,"statistic":117,"alignment":118},"median relative pose error (RPE) in mm","mm","median","not_reported",[120,124,126,128,130,132,134,136,138,140,142,144,146,148,150,152,154,156,158,160,162,164],{"dataset":121,"sequence":122,"environment":123},"TUM RGB-D (Freiburg)","fr1 360","indoor office and structure\u002Ftexture test scenes, RGB-D camera (carrying mode not stated in the paper)",{"dataset":121,"sequence":125,"environment":123},"fr1 desk",{"dataset":121,"sequence":127,"environment":123},"fr1 desk2",{"dataset":121,"sequence":129,"environment":123},"fr1 floor",{"dataset":121,"sequence":131,"environment":123},"fr1 plant",{"dataset":121,"sequence":133,"environment":123},"fr1 room",{"dataset":121,"sequence":135,"environment":123},"fr1 rpy",{"dataset":121,"sequence":137,"environment":123},"fr1 teddy",{"dataset":121,"sequence":139,"environment":123},"fr1 xyz",{"dataset":121,"sequence":141,"environment":123},"fr2 desk",{"dataset":121,"sequence":143,"environment":123},"fr2 large no loop",{"dataset":121,"sequence":145,"environment":123},"fr2 rpy",{"dataset":121,"sequence":147,"environment":123},"fr2 xyz",{"dataset":121,"sequence":149,"environment":123},"fr3 long office household",{"dataset":121,"sequence":151,"environment":123},"fr3 nostruct. notext. far",{"dataset":121,"sequence":153,"environment":123},"fr3 nostruct. notext. near",{"dataset":121,"sequence":155,"environment":123},"fr3 nostruct. text. far",{"dataset":121,"sequence":157,"environment":123},"fr3 nostruct. text. near",{"dataset":121,"sequence":159,"environment":123},"fr3 struct. notext. far",{"dataset":121,"sequence":161,"environment":123},"fr3 struct. notext. near",{"dataset":121,"sequence":163,"environment":123},"fr3 struct. text. far",{"dataset":121,"sequence":165,"environment":123},"fr3 struct. text. near",[167,169,171,174,177],{"name":168,"methodId":5,"linkable":80,"proposed":80,"self":80},"Ours (MRSMap)",{"name":170,"methodId":70,"linkable":76,"proposed":76,"self":76},"Warp [17] (OpenCV)",{"name":172,"methodId":173,"linkable":80,"proposed":76,"self":76},"GICP [5]","segal2009gicp",{"name":175,"methodId":176,"linkable":80,"proposed":76,"self":76},"3D-NDT [7]","stoyanov2012d2dndt",{"name":178,"methodId":179,"linkable":80,"proposed":76,"self":76},"Fovis [12]","fovis2017",[181,185,188,191,194,197,199,201,203,205,207,209,211,213,215,217,219,221,223,224,226,228,230,232,234,236,237,238,239,241,243,245,246,247,249,250,251,252,254,256,257,259,261,263,265,266,268,269,271,273,275,278,280,281,283,285,287,289,291,292,293,296,297,298,299,301,303,305,306,307,309,312,314,316,318,320,323,325,327,329,331,334,336,338,340,342,345,346,348,350,352,355,356,357,359,361,363,364,365,367,369,372,374,375,376,377,379,380,382,383],[182,182,182,183,184,182,184,184,182],0,5.1,-1,[186,182,182,187,184,182,184,184,182],1,5.9,[189,182,182,190,184,182,184,184,182],2,18.8,[192,182,182,193,184,182,184,184,182],3,7.8,[195,182,182,196,184,182,184,184,182],4,7.1,[182,182,186,198,184,182,184,184,182],4.4,[186,182,186,200,184,182,184,184,182],5.8,[189,182,186,202,184,182,184,184,182],10.2,[192,182,186,204,184,182,184,184,182],7.9,[195,182,186,206,184,182,184,184,182],6.3,[182,182,189,208,184,182,184,184,182],4.5,[186,182,189,210,184,182,184,184,182],6.2,[189,182,189,212,184,182,184,184,182],10.4,[192,182,189,214,184,182,184,184,182],8.2,[195,182,189,216,184,182,184,184,182],6.6,[182,182,192,218,184,182,184,184,182],4.9,[186,182,192,220,184,182,184,184,182],2.1,[189,182,192,222,184,182,184,184,182],5,[192,182,192,206,184,182,184,184,182],[195,182,192,225,184,182,184,184,182],2.6,[182,182,195,227,184,182,184,184,182],3.5,[186,182,195,229,184,182,184,184,182],4.2,[189,182,195,231,184,182,184,184,182],16.1,[192,182,195,233,184,182,184,184,182],7.4,[195,182,195,235,184,182,184,184,182],4.6,[182,182,222,227,184,182,184,184,182],[186,182,222,235,184,182,184,184,182],[189,182,222,202,184,182,184,184,182],[192,182,222,240,184,182,184,184,182],6.1,[195,182,222,242,184,182,184,184,182],5.4,[182,182,244,192,184,182,184,184,182],6,[186,182,244,183,184,182,184,184,182],[189,182,244,212,184,182,184,184,182],[192,182,244,248,184,182,184,184,182],6.8,[195,182,244,242,184,182,184,184,182],[182,182,106,229,184,182,184,184,182],[186,182,106,240,184,182,184,184,182],[189,182,106,253,184,182,184,184,182],21.3,[192,182,106,255,184,182,184,184,182],8.8,[195,182,106,196,184,182,184,184,182],[182,182,258,225,184,182,184,184,182],8,[186,182,258,260,184,182,184,184,182],4.1,[189,182,258,262,184,182,184,184,182],3.9,[192,182,258,264,184,182,184,184,182],5.2,[195,182,258,235,184,182,184,184,182],[182,182,267,220,184,182,184,184,182],9,[186,182,267,220,184,182,184,184,182],[189,182,267,270,184,182,184,184,182],6.7,[192,182,267,272,184,182,184,184,182],4.3,[195,182,267,274,184,182,184,184,182],2.5,[182,182,276,277,184,182,184,184,182],10,21.8,[186,182,276,279,184,182,184,184,182],20.5,[189,182,276,253,184,182,184,184,182],[192,182,276,282,184,182,184,184,182],32.1,[195,182,276,284,184,182,184,184,182],11,[182,182,284,286,184,182,184,184,182],1.6,[186,182,284,288,184,182,184,184,182],1.7,[189,182,284,290,184,182,184,184,182],1.3,[192,182,284,229,184,182,184,184,182],[195,182,284,288,184,182,184,184,182],[182,182,294,295,184,182,184,184,182],12,1.4,[186,182,294,189,184,182,184,184,182],[189,182,294,288,184,182,184,184,182],[192,182,294,195,184,182,184,184,182],[195,182,294,300,184,182,184,184,182],1.9,[182,182,302,225,184,182,184,184,182],13,[186,182,302,304,184,182,184,184,182],3.2,[189,182,302,193,184,182,184,184,182],[192,182,302,229,184,182,184,184,182],[195,182,302,308,184,182,184,184,182],3.7,[182,182,310,311,184,182,184,184,182],14,9.7,[186,182,310,313,184,182,184,184,182],40.4,[189,182,310,315,184,182,184,184,182],8.6,[192,182,310,317,184,182,184,184,182],13.8,[195,182,310,319,184,182,184,184,182],11.3,[182,182,321,322,184,182,184,184,182],15,15.2,[186,182,321,324,184,182,184,184,182],28.2,[189,182,321,326,184,182,184,184,182],12.5,[192,182,321,328,184,182,184,184,182],17.1,[195,182,321,330,184,182,184,184,182],11.2,[182,182,332,333,184,182,184,184,182],16,18.5,[186,182,332,335,184,182,184,184,182],19.2,[189,182,332,337,184,182,184,184,182],10.9,[192,182,332,339,184,182,184,184,182],18.6,[195,182,332,341,184,182,184,184,182],20.8,[182,182,343,344,184,182,184,184,182],17,11.5,[186,182,343,106,184,182,184,184,182],[189,182,343,347,184,182,184,184,182],8.9,[192,182,343,349,184,182,184,184,182],10.6,[195,182,343,351,184,182,184,184,182],7.3,[182,182,353,354,184,182,184,184,182],18,2.2,[186,182,353,315,184,182,184,184,182],[189,182,353,208,184,182,184,184,182],[192,182,353,358,184,182,184,184,182],2.9,[195,182,353,360,184,182,184,184,182],9.1,[182,182,362,220,184,182,184,184,182],19,[186,182,362,315,184,182,184,184,182],[189,182,362,358,184,182,184,184,182],[192,182,362,366,184,182,184,184,182],2.4,[195,182,362,368,184,182,184,184,182],9.3,[182,182,370,371,184,182,184,184,182],20,5.5,[186,182,370,373,184,182,184,184,182],8.1,[189,182,370,196,184,182,184,184,182],[192,182,370,242,184,182,184,184,182],[195,182,370,255,184,182,184,184,182],[182,182,378,304,184,182,184,184,182],21,[186,182,378,187,184,182,184,184,182],[189,182,378,381,184,182,184,184,182],5.6,[192,182,378,371,184,182,184,184,182],[195,182,378,384,184,182,184,184,182],6.5,[],[111],[],[],[390],"Incremental (frame-to-frame) registration on TUM Freiburg sequences; median translational relative pose error in mm (maximum values in brackets in the table not extracted); warp is the OpenCV reimplementation",{"slug":392,"group":393,"sourceId":5,"sourceLabel":6,"table":394,"selfRows":112,"metrics":395,"seqs":400,"entrants":423,"cells":430,"outcomes":497,"locators":498,"hardware":500,"wordings":501,"notes":502},"mrsmap2014-table-3-corrigendum","mrsmap2014:Table 3 (corrigendum)","Table 3 (corrigendum)",[396],{"label":397,"unit":398,"statistic":399,"alignment":118},"RMSE RPE in m","m","RMSE",[401,403,405,407,409,411,413,415,417,419,421],{"dataset":121,"sequence":402,"environment":123},"freiburg1_360",{"dataset":121,"sequence":404,"environment":123},"freiburg1_desk2",{"dataset":121,"sequence":406,"environment":123},"freiburg1_desk",{"dataset":121,"sequence":408,"environment":123},"freiburg1_plant",{"dataset":121,"sequence":410,"environment":123},"freiburg1_room",{"dataset":121,"sequence":412,"environment":123},"freiburg1_rpy",{"dataset":121,"sequence":414,"environment":123},"freiburg1_teddy",{"dataset":121,"sequence":416,"environment":123},"freiburg1_xyz",{"dataset":121,"sequence":418,"environment":123},"freiburg2_desk",{"dataset":121,"sequence":420,"environment":123},"freiburg2_rpy",{"dataset":121,"sequence":422,"environment":123},"freiburg2_xyz",[424,426,428],{"name":425,"methodId":5,"linkable":80,"proposed":80,"self":80},"Ours all frames (MRSMap SLAM)",{"name":427,"methodId":5,"linkable":80,"proposed":80,"self":80},"Ours real-time (MRSMap SLAM, frames dropped, 0.05 m max. resolution)",{"name":429,"methodId":70,"linkable":76,"proposed":76,"self":76},"RGB-D SLAM [24], [34] (Endres et al.)",[431,433,435,437,439,441,443,445,447,449,451,453,455,457,459,461,463,465,467,469,471,473,475,477,479,481,483,485,487,489,491,493,495],[182,182,182,432,184,182,184,184,182],0.123,[186,182,182,434,184,182,184,184,182],0.126,[189,182,182,436,184,182,184,184,182],0.103,[182,182,186,438,184,182,184,184,182],0.098,[186,182,186,440,184,182,184,184,182],0.107,[189,182,186,442,184,182,184,184,182],0.102,[182,182,189,444,184,182,184,184,182],0.054,[186,182,189,446,184,182,184,184,182],0.088,[189,182,189,448,184,182,184,184,182],0.049,[182,182,192,450,184,182,184,184,182],0.038,[186,182,192,452,184,182,184,184,182],0.062,[189,182,192,454,184,182,184,184,182],0.142,[182,182,195,456,184,182,184,184,182],0.111,[186,182,195,458,184,182,184,184,182],0.145,[189,182,195,460,184,182,184,184,182],0.219,[182,182,222,462,184,182,184,184,182],0.041,[186,182,222,464,184,182,184,184,182],0.045,[189,182,222,466,184,182,184,184,182],0.042,[182,182,244,468,184,182,184,184,182],0.066,[186,182,244,470,184,182,184,184,182],0.092,[189,182,244,472,184,182,184,184,182],0.138,[182,182,106,474,184,182,184,184,182],0.02,[186,182,106,476,184,182,184,184,182],0.025,[189,182,106,478,184,182,184,184,182],0.021,[182,182,258,480,184,182,184,184,182],0.1,[186,182,258,482,184,182,184,184,182],0.115,[189,182,258,484,184,182,184,184,182],0.143,[182,182,267,486,184,182,184,184,182],0.043,[186,182,267,488,184,182,184,184,182],0.04,[189,182,267,490,184,182,184,184,182],0.026,[182,182,276,492,184,182,184,184,182],0.032,[186,182,276,494,184,182,184,184,182],0.033,[189,182,276,496,184,182,184,184,182],0.037,[],[499],"Corrigendum Table 3 (replaces article Table 3)",[],[],[503],"SLAM on TUM Freiburg sequences: RMSE of relative pose error averaged over all frame differences, in m; values from the 2015 corrigendum that replaces the originally printed Table 3",{"slug":505,"group":506,"sourceId":507,"sourceLabel":508,"table":509,"selfRows":195,"metrics":510,"seqs":513,"entrants":524,"cells":541,"outcomes":587,"locators":589,"hardware":590,"wordings":591,"notes":592},"psmslam2017-table-2","psmslam2017:Table 2","psmslam2017","Yan et al., 2017","Table 2",[511],{"label":512,"unit":398,"statistic":399,"alignment":118},"ATE [m]",[514,518,520,522],{"dataset":515,"sequence":516,"environment":517},"TUM RGB-D","fr1\u002Fdesk","indoor office scenes, RGB-D camera (carrying mode not stated in the paper)",{"dataset":515,"sequence":519,"environment":517},"fr3\u002Foffice",{"dataset":515,"sequence":521,"environment":517},"fr1\u002F360",{"dataset":515,"sequence":523,"environment":517},"fr1\u002Froom",[525,527,530,532,535,537,539],{"name":526,"methodId":70,"linkable":76,"proposed":76,"self":76},"RGB-D SLAM [4] (Endres et al.)",{"name":528,"methodId":529,"linkable":80,"proposed":76,"self":76},"Kintinuous [33]","kintinuous2015",{"name":531,"methodId":5,"linkable":80,"proposed":76,"self":80},"MRSMap [28]",{"name":533,"methodId":534,"linkable":80,"proposed":76,"self":76},"ElasticFusion [34]","elasticfusion2015",{"name":536,"methodId":70,"linkable":76,"proposed":76,"self":76},"DVO SLAM [14]",{"name":538,"methodId":70,"linkable":76,"proposed":76,"self":76},"sigma-DVO SLAM",{"name":540,"methodId":507,"linkable":80,"proposed":80,"self":76},"PSM SLAM",[542,544,545,546,547,548,550,552,553,555,556,558,560,562,564,566,567,569,570,572,574,576,578,579,580,581,583,585],[182,182,182,543,184,182,184,184,182],0.023,[186,182,182,496,184,182,184,184,182],[189,182,182,486,184,182,184,184,182],[192,182,182,474,184,182,184,184,182],[195,182,182,478,184,182,184,184,182],[222,182,182,549,184,182,184,184,182],0.019,[244,182,182,551,184,182,184,184,182],0.016,[182,182,186,492,184,182,184,184,182],[186,182,186,554,184,182,184,184,182],0.03,[189,182,186,466,184,182,184,184,182],[192,182,186,557,184,182,184,184,182],0.017,[195,182,186,559,184,182,184,184,182],0.035,[222,182,186,561,184,182,184,184,182],0.015,[244,182,186,563,184,182,184,184,182],0.031,[182,182,189,565,184,182,184,184,182],0.079,[186,182,189,70,182,182,184,184,182],[189,182,189,568,184,182,184,184,182],0.069,[192,182,189,70,182,182,184,184,182],[195,182,189,571,184,182,184,184,182],0.083,[222,182,189,573,184,182,184,184,182],0.061,[244,182,189,575,184,182,184,184,182],0.055,[182,182,192,577,184,182,184,184,182],0.084,[186,182,192,70,182,182,184,184,182],[189,182,192,568,184,182,184,184,182],[192,182,192,70,182,182,184,184,182],[195,182,192,582,184,182,184,184,182],0.053,[222,182,192,584,184,182,184,184,182],0.06,[244,182,192,586,184,182,184,184,182],0.051,[588],"not reported",[509],[],[],[593],"Complete SLAM absolute trajectory error on TUM RGB-D; '-' = no result given",{"slug":595,"group":596,"sourceId":507,"sourceLabel":508,"table":597,"selfRows":195,"metrics":598,"seqs":602,"entrants":613,"cells":620,"outcomes":658,"locators":659,"hardware":660,"wordings":661,"notes":662},"psmslam2017-table-4","psmslam2017:Table 4","Table 4",[599],{"label":600,"unit":398,"statistic":601,"alignment":118},"mean distance from points to nearest ground-truth surface (m)","mean",[603,607,609,611],{"dataset":604,"sequence":605,"environment":606},"ICL-NUIM","lr kt0","synthetic living room",{"dataset":604,"sequence":608,"environment":606},"lr kt1",{"dataset":604,"sequence":610,"environment":606},"lr kt2",{"dataset":604,"sequence":612,"environment":606},"lr kt3",[614,615,616,617,618,619],{"name":526,"methodId":70,"linkable":76,"proposed":76,"self":76},{"name":528,"methodId":529,"linkable":80,"proposed":76,"self":76},{"name":531,"methodId":5,"linkable":80,"proposed":76,"self":80},{"name":536,"methodId":70,"linkable":76,"proposed":76,"self":76},{"name":533,"methodId":534,"linkable":80,"proposed":76,"self":76},{"name":540,"methodId":507,"linkable":80,"proposed":80,"self":76},[621,623,625,626,627,629,631,632,634,636,637,638,640,641,642,643,645,646,648,650,652,654,655,657],[182,182,182,622,184,182,184,184,182],0.044,[186,182,182,624,184,182,184,184,182],0.011,[189,182,182,573,184,182,184,184,182],[192,182,182,492,184,182,184,184,182],[195,182,182,628,184,182,184,184,182],0.007,[222,182,182,630,184,182,184,184,182],0.006,[182,182,186,492,184,182,184,184,182],[186,182,186,633,184,182,184,184,182],0.008,[189,182,186,635,184,182,184,184,182],0.14,[192,182,186,573,184,182,184,184,182],[195,182,186,628,184,182,184,184,182],[222,182,186,639,184,182,184,184,182],0.009,[182,182,189,563,184,182,184,184,182],[186,182,189,639,184,182,184,184,182],[189,182,189,438,184,182,184,184,182],[192,182,189,644,184,182,184,184,182],0.119,[195,182,189,633,184,182,184,184,182],[222,182,189,647,184,182,184,184,182],0.024,[182,182,192,649,184,182,184,184,182],0.167,[186,182,192,651,184,182,184,184,182],0.15,[189,182,192,653,184,182,184,184,182],0.248,[192,182,192,582,184,182,184,184,182],[195,182,192,656,184,182,184,184,182],0.028,[222,182,192,639,184,182,184,184,182],[],[597],[],[],[663],"Surface reconstruction accuracy on ICL-NUIM living room with noise: mean distance from reconstructed points to the nearest ground-truth surface (m)",[665,670,674],{"group":666,"slug":667,"sourceLabel":508,"table":668,"selfRows":195,"datasets":669},"psmslam2017:Table 5","psmslam2017-table-5","Table 5",[604],{"group":671,"slug":672,"sourceLabel":6,"table":509,"selfRows":189,"datasets":673},"mrsmap2014:Table 2","mrsmap2014-table-2",[121],{"group":675,"slug":676,"sourceLabel":6,"table":677,"selfRows":189,"datasets":678},"mrsmap2014:Text Sec.6.2","mrsmap2014-text-sec-6-2","Text Sec.6.2",[121],1790510661841]