[{"data":1,"prerenderedAt":674},["ShallowReactive",2],{"method-infinitam2015":3},{"method":4,"reference":65,"equipment":90,"figures":128,"results":129},{"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":28,"sensors":34,"platform":38,"estimator":42,"association":43,"timeModel":44,"deskew":45,"loopClosure":46,"globalOptimization":47,"mapRepresentation":48,"prior":47,"outputGeometry":49,"compute":50,"codeUrl":51,"codeLicense":52,"relatedVersions":53},"infinitam2015","Kähler et al., 2015","InfiniTAM","Very High Frame Rate Volumetric Integration of Depth Images on Mobile Devices",2015,"classic","C08","odometry_with_local_mapping","本文把 KinectFusion 式的 TSDF 稠密重建最佳化到能在平板電腦上即時執行。資料結構沿用 Nießner 等人的體素區塊雜湊（每塊 8×8×8 體素），但改成每個桶只有一個表頭、碰撞放入額外鏈結串列的雜湊表，並以免鎖的兩步配置、只檢查上一影格可見區塊的增量可見清單，以及固定大小傳輸緩衝的 GPU 與主記憶體（或磁碟）資料交換，控制每一影格的運算量與延遲。射線投射先以低解析度的區塊投影求出搜尋範圍，並可在相機移動不大時跳過部分影格的射線投射；追蹤採點對面 ICP 或色彩直接對齊，也可直接用平板 IMU 提供旋轉，以減少旋轉漂移。整套程式以 InfiniTAM 框架公開。","Very high frame rate TSDF fusion on mobile devices: voxel-block hashing with a chained single-entry hash table, lock-free allocation, incremental visible-block lists, bounded host swapping, coarse raycast bounds and optional raycast skipping, with ICP or colour tracking whose rotation can come from the tablet IMU; released as the InfiniTAM framework.","full_text_reviewed","peer_reviewed_published","main_body","論文未在施工現場測試；實驗為桌面、客廳與 ICL-NUIM 合成室內，並以定性圖展示較大空間的重建（Fig. 16）。以平板電腦搭配 Structure Sensor 與 IMU 即時重建，符合低成本手持室內掃描的需求；但系統沒有迴圈閉合，作者指出大尺度重建必然累積漂移，後續 InfiniTAM v3 報告（arXiv 1708.00783，依其題名）才加入迴圈閉合。",[20,21,22],"public_benchmark","simulation","controlled_experiment",[24,25,26,27],"47 Hz on an Nvidia Shield Tablet with 320x240 depth and IMU, about 20 Hz on an iPad Air 2, and up to 910 Hz with visualisation (over 1.1 kHz without) on a GTX Titan X (Sec. 1; Table 1)","About an order of magnitude faster than the KinectFusion and voxel hashing implementations on the same GPU (teddy: 1.91 ms versus 26.15 ms and 25.87 ms) (Table 1)","IMU rotation kept drift over a full swivel-chair rotation to at most 0.091 deg, versus up to 23 deg with ICP (Sec. 7.4; Fig. 14)","ICL-NUIM trajectory and surface errors lower than the best values of the ICL-NUIM benchmark paper on most sequences, including with approximate raycasting (Tables 3 to 5)",[29,30,31,32,33],"No loop closure or pose-graph optimization; truly large-scale reconstructions accumulate drift (Sec. 8)","Intra-frame hash collisions during allocation are resolved arbitrarily and may drop blocks until the next frame (Sec. 2.2)","Image-space normals are inaccurate at depth discontinuities (Sec. 4.3)","ICP drift in two ICL-NUIM cases caused by non-informative depth (Fig. 15)","The fixed-volume KinectFusion comparison could not cover the whole couch scene, so runtime comparisons use different scene coverage (Sec. 7.1) (inference)",[35,36,37],"Depth camera: Microsoft Kinect for XBOX 360 (teddy sequence, 640x480 colour and disparity)","Depth camera: Occipital Structure Sensor (couch sequence, 320x240 depth)","IMU of the tablet (Apple iPad Air 2 orientation for the couch sequence; tablet IMU in the swivel-chair test)",[39,40,41],"recorded sequences: teddy desk scene (Kinect for XBOX 360) and couch living-room scene (Structure Sensor with iPad Air 2 IMU); carrying mode not stated (Sec. 7.1, Table 1)","tablet fixed to a swivel chair (IMU rotation-drift test, Sec. 7.4)","simulation (ICL-NUIM living room and office sequences, Sec. 7.5)","Frame-to-model point-to-plane ICP (or colour-based direct image alignment) solved by Gauss-Newton on resolution hierarchies, rotation only at coarse levels; optionally the rotation is taken from the tablet IMU via an inertial fusion algorithm and only translation is estimated visually","Projective association between the current depth image and point and normal maps ray-cast from the TSDF (normals computed in image space)","discrete poses","not_applicable (depth camera)","none (stated as not addressed; loop closure is future work)","none","TSDF stored in 8x8x8 voxel blocks addressed by a hash table with one list-head entry per bucket plus an excess linked list; incremental visible-block list; blocks swapped between device memory and host memory or disk through fixed-size transfer buffers","TSDF voxel-block map with raycast point clouds and renderings; optional colour per voxel","GPU (Nvidia CUDA), tablet APU or multi-core CPU; 1.10 to 1.91 ms per frame on a GTX Titan X (up to 910 Hz with visualisation, beyond 1.1 kHz without), 21.04 ms (47 Hz) on an Nvidia Shield Tablet (Tegra K1) with 320x240 depth and IMU, about 20 Hz on an Apple iPad Air 2 (Table 1, Sec. 1)","https:\u002F\u002Fgithub.com\u002Fvictorprad\u002FInfiniTAM","custom Oxford University Innovation (Isis Innovation) licence, non-commercial use only (LICENSE file checked)",[54,58,62],{"relation":55,"title":56,"doi_or_url":57},"related_technical_report","A Framework for the Volumetric Integration of Depth Images (InfiniTAM framework report, arXiv 1410.0925; not the TVCG paper)","https:\u002F\u002Farxiv.org\u002Fabs\u002F1410.0925",{"relation":59,"title":60,"doi_or_url":61},"successor_preprint","InfiniTAM v3: A Framework for Large-Scale 3D Reconstruction with Loop Closure (arXiv 1708.00783)","https:\u002F\u002Farxiv.org\u002Fabs\u002F1708.00783",{"relation":63,"title":64,"doi_or_url":51},"code_release","victorprad\u002FInfiniTAM (the paper points to www.infinitam.org)",{"id":5,"kind":66,"shortName":7,"title":8,"authors":67,"year":9,"venue":74,"venueType":75,"publisher":76,"volumeIssuePages":77,"doi":78,"arxivId":79,"url":80,"firstPublicDate":81,"publicationStatus":16,"metadataStatus":82,"fulltextStatus":15,"era":10,"classicReason":83,"codeUrl":51,"cluster":11,"topics":84,"mdpi":85,"verification":86,"label":6,"fulltextRoute":87,"versionRead":88,"addedByCensus":89},"method",[68,69,70,71,72,73],"Olaf Kähler","Victor Adrian Prisacariu","Carl Yuheng Ren","Xin Sun","Philip Torr","David Murray","IEEE Transactions on Visualization and Computer Graphics (ISMAR 2015 special issue)","journal","IEEE","21(11):1241-1250","10.1109\u002Ftvcg.2015.2459891",null,"https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1109\u002FTVCG.2015.2459891","2015-07-23","metadata_verified","principle reused: voxel-block hashing re-engineered for very high frame rates (chained single-entry hash table, lock-free allocation, incremental visible-block list, bounded host swapping, coarse raycast bounds) so that TSDF fusion runs on tablets; the open InfiniTAM framework became a standard volumetric baseline (compared in supereight2018 and surfelmeshing2020; flashfusion2018 compares against the later ECCV 2016 InfiniTAM with loop closure, its ref. [10]) and bridges voxelhashing2013 and later GPU TSDF libraries such as millane2024nvblox.",[11],false,"corrected","NTU institutional (Chrome)","IEEE Xplore HTML full text of the version of record (TVCG 21(11), 2015) with Tables 1 to 5 viewed as publisher images",true,[91,97,100,105,111,114,119,123],{"category":92,"model":93,"canonical":93,"role":94,"dataset":79,"specs":95,"locator":96},"rgbd","Microsoft Kinect for XBOX 360","method input","colour and disparity images at 640x480 (teddy sequence)","Sec. 7.1",{"category":92,"model":98,"canonical":98,"role":94,"dataset":79,"specs":99,"locator":96},"Occipital Structure Sensor","depth images at 320x240 (couch sequence)",{"category":101,"model":102,"canonical":102,"role":94,"dataset":79,"specs":103,"locator":104},"imu","IMU of an Apple iPad Air 2","orientation information for the couch sequence; rotation replaces visual rotation estimation","Sec. 5; Sec. 7.1",{"category":106,"model":107,"canonical":107,"role":108,"dataset":79,"specs":109,"locator":110},"mobile_scanner_device","Nvidia Shield Tablet (Nvidia Tegra K1)","compute for runtime","runs the full pipeline at up to 47 Hz on 320x240 depth with IMU","Sec. 1; Table 1",{"category":106,"model":112,"canonical":112,"role":108,"dataset":79,"specs":113,"locator":110},"Apple iPad Air 2","runs the full pipeline at about 20 Hz",{"category":115,"model":116,"canonical":116,"role":108,"dataset":79,"specs":117,"locator":118},"compute","Nvidia GTX Titan X GPU","up to 910 Hz with visualisation, beyond 1.1 kHz without","Sec. 1; Table 1; Fig. 12",{"category":115,"model":120,"canonical":120,"role":108,"dataset":79,"specs":121,"locator":122},"Intel Core i7-5960X (CPU-only implementation, partly OpenMP)","workstation CPU runs","Sec. 7.1; Table 1",{"category":124,"model":125,"canonical":125,"role":94,"dataset":79,"specs":126,"locator":127},"other","tablet fixed to a swivel chair (tablet model not stated)","full rotation of the chair to measure rotation drift of ICP versus IMU tracking","Sec. 7.4; Fig. 13",[],{"totalRows":130,"groupCount":131,"groups":132,"others":653},102,8,[133,230,447,548],{"slug":134,"group":135,"sourceId":5,"sourceLabel":6,"table":136,"selfRows":137,"metrics":138,"seqs":148,"entrants":157,"cells":164,"outcomes":218,"locators":219,"hardware":222,"wordings":226,"notes":227},"infinitam2015-table-1","infinitam2015:Table 1","Table 1",24,[139,144,146],{"label":140,"unit":141,"statistic":142,"alignment":143},"average computation time per frame (full)","ms","mean","not_reported",{"label":145,"unit":141,"statistic":142,"alignment":143},"average computation time per frame (forward)",{"label":147,"unit":141,"statistic":142,"alignment":143},"average computation time per frame (none)",[149,153],{"dataset":150,"sequence":151,"environment":152},"authors' teddy sequence","teddy","indoor desk scene",{"dataset":154,"sequence":155,"environment":156},"authors' couch sequence","couch","indoor living room, Structure Sensor with iPad Air 2 IMU (carrying mode not stated)",[158,160,162],{"name":159,"methodId":5,"linkable":89,"proposed":89,"self":89},"InfiniTAM (full raycast every frame)",{"name":161,"methodId":5,"linkable":89,"proposed":89,"self":89},"InfiniTAM (forward projection)",{"name":163,"methodId":5,"linkable":89,"proposed":89,"self":89},"InfiniTAM (no visualisation)",[165,169,172,175,177,179,181,183,185,187,190,192,194,196,198,200,202,204,206,208,210,212,214,216],[166,166,166,167,168,166,166,168,166],0,1.91,-1,[170,170,166,171,168,166,166,168,166],1,1.74,[173,173,166,174,168,166,166,168,166],2,1.38,[166,166,166,176,168,166,170,168,166],36.53,[170,170,166,178,168,166,170,168,166],31.38,[173,173,166,180,168,166,170,168,166],26.79,[166,166,166,182,168,166,173,168,166],82.6,[170,170,166,184,168,166,173,168,166],65.55,[173,173,166,186,168,166,173,168,166],56.1,[166,166,166,188,168,166,189,168,166],45.28,3,[170,170,166,191,168,166,189,168,166],46.75,[173,173,166,193,168,166,189,168,166],35.4,[166,166,170,195,168,170,166,168,170],1.17,[170,170,170,197,168,170,166,168,170],1.1,[173,173,170,199,168,170,166,168,170],0.87,[166,166,170,201,168,170,170,168,170],25.58,[170,170,170,203,168,170,170,168,170],21.04,[173,173,170,205,168,170,170,168,170],19.38,[166,166,170,207,168,170,173,168,170],56.65,[170,170,170,209,168,170,173,168,170],48.43,[173,173,170,211,168,170,173,168,170],41.58,[166,166,170,213,168,170,189,168,170],23.43,[170,170,170,215,168,170,189,168,170],23.38,[173,173,170,217,168,170,189,168,170],19.94,[],[220,221],"Table 1(a)","Table 1(b)",[223,224,112,225],"Nvidia Titan X","Nvidia Tegra K1","Intel Core i7-5960X",[],[228,229],"Average computation time per frame over the teddy sequence (Kinect for XBOX 360, 640x480 colour and disparity, no IMU) for three visualisation strategies of InfiniTAM and for the KinectFusion [14] and voxel hashing [16] implementations","Average computation time per frame over the couch sequence (Structure Sensor 320x240 depth with iPad Air 2 IMU) for three visualisation strategies of InfiniTAM and for KinectFusion [14] and voxel hashing [16]; the KinectFusion fixed 512^3 volume could not cover the whole couch scene",{"slug":231,"group":232,"sourceId":233,"sourceLabel":234,"table":235,"selfRows":137,"metrics":236,"seqs":246,"entrants":257,"cells":273,"outcomes":440,"locators":441,"hardware":443,"wordings":444,"notes":445},"surfelmeshing2020-table-2-ground-truth-trajectories","surfelmeshing2020:Table 2 (ground-truth trajectories)","surfelmeshing2020","Schöps et al., 2020","Table 2 (ground-truth trajectories)",[237,241,243],{"label":238,"unit":239,"statistic":143,"alignment":240},"Accuracy [%] (share of reconstructed surfels within 1 cm of ground truth)","%","SE3",{"label":242,"unit":239,"statistic":143,"alignment":240},"Completeness [%] (share of ground-truth points within 1 cm of the reconstruction)",{"label":244,"unit":245,"statistic":142,"alignment":240},"Curvature [0.01\u002Fm] (mean curvature, smoothness)","0.01 1\u002Fm",[247,251,253,255],{"dataset":248,"sequence":249,"environment":250},"ICL-NUIM","kt0","synthetic indoor living room",{"dataset":248,"sequence":252,"environment":250},"kt1",{"dataset":248,"sequence":254,"environment":250},"kt2",{"dataset":248,"sequence":256,"environment":250},"kt3",[258,260,262,264,266,269,271],{"name":259,"methodId":5,"linkable":89,"proposed":85,"self":89},"InfiniTAM [29]",{"name":261,"methodId":5,"linkable":89,"proposed":85,"self":89},"InfiniTAM [29] - smoothed",{"name":263,"methodId":79,"linkable":85,"proposed":85,"self":85},"FastFusion [27]",{"name":265,"methodId":79,"linkable":85,"proposed":85,"self":85},"FastFusion [27] - smoothed",{"name":267,"methodId":268,"linkable":89,"proposed":85,"self":85},"ElasticFusion [17]","elasticfusion2015",{"name":270,"methodId":268,"linkable":89,"proposed":85,"self":85},"ElasticFusion [17] - smoothed",{"name":272,"methodId":233,"linkable":89,"proposed":89,"self":85},"SurfelMeshing (Ours)",[274,276,278,280,282,285,288,291,293,295,297,299,301,303,305,307,309,311,313,315,317,319,321,323,325,327,329,331,333,335,337,339,341,343,345,346,348,350,352,354,356,358,360,362,364,366,367,369,371,373,375,377,379,380,382,384,386,388,390,392,394,396,398,400,402,404,406,407,409,411,413,415,417,419,421,423,425,426,428,430,432,434,436,438],[166,166,166,275,168,166,168,168,166],76.4,[170,166,166,277,168,166,168,168,166],78.3,[173,166,166,279,168,166,168,168,166],85.5,[189,166,166,281,168,166,168,168,166],75.9,[283,166,166,284,168,166,168,168,166],4,96.2,[286,166,166,287,168,166,168,168,166],5,95.7,[289,166,166,290,168,166,168,168,166],6,93.5,[166,166,170,292,168,166,168,168,166],68.4,[170,166,170,294,168,166,168,168,166],68.3,[173,166,170,296,168,166,168,168,166],80.1,[189,166,170,298,168,166,168,168,166],78.8,[283,166,170,300,168,166,168,168,166],83.1,[286,166,170,302,168,166,168,168,166],82.1,[289,166,170,304,168,166,168,168,166],86.4,[166,166,173,306,168,166,168,168,166],56.6,[170,166,173,308,168,166,168,168,166],58.1,[173,166,173,310,168,166,168,168,166],64.2,[189,166,173,312,168,166,168,168,166],52.6,[283,166,173,314,168,166,168,168,166],97.1,[286,166,173,316,168,166,168,168,166],96.6,[289,166,173,318,168,166,168,168,166],69.5,[166,166,189,320,168,166,168,168,166],54.4,[170,166,189,322,168,166,168,168,166],58.3,[173,166,189,324,168,166,168,168,166],85.3,[189,166,189,326,168,166,168,168,166],72.5,[283,166,189,328,168,166,168,168,166],93.7,[286,166,189,330,168,166,168,168,166],92.2,[289,166,189,332,168,166,168,168,166],74,[166,170,166,334,168,166,168,168,166],53.5,[170,170,166,336,168,166,168,168,166],51.6,[173,170,166,338,168,166,168,168,166],54.7,[189,170,166,340,168,166,168,168,166],45.6,[283,170,166,342,168,166,168,168,166],38.8,[286,170,166,344,168,166,168,168,166],38.9,[289,170,166,340,168,166,168,168,166],[166,170,170,347,168,166,168,168,166],66.5,[170,170,170,349,168,166,168,168,166],62.8,[173,170,170,351,168,166,168,168,166],67.3,[189,170,170,353,168,166,168,168,166],63,[283,170,170,355,168,166,168,168,166],46,[286,170,170,357,168,166,168,168,166],45.3,[289,170,170,359,168,166,168,168,166],58.6,[166,170,173,361,168,166,168,168,166],46.7,[170,170,173,363,168,166,168,168,166],42.6,[173,170,173,365,168,166,168,168,166],48.4,[189,170,173,342,168,166,168,168,166],[283,170,173,368,168,166,168,168,166],22.8,[286,170,173,370,168,166,168,168,166],23.1,[289,170,173,372,168,166,168,168,166],30.8,[166,170,189,374,168,166,168,168,166],61.6,[170,170,189,376,168,166,168,168,166],59.9,[173,170,189,378,168,166,168,168,166],82.8,[189,170,189,347,168,166,168,168,166],[283,170,189,381,168,166,168,168,166],40,[286,170,189,383,168,166,168,168,166],40.4,[289,170,189,385,168,166,168,168,166],52.1,[166,173,166,387,168,166,168,168,166],2.48,[170,173,166,389,168,166,168,168,166],1.46,[173,173,166,391,168,166,168,168,166],0.99,[189,173,166,393,168,166,168,168,166],0.63,[283,173,166,395,168,166,168,168,166],0.24,[286,173,166,397,168,166,168,168,166],0.18,[289,173,166,399,168,166,168,168,166],0.15,[166,173,170,401,168,166,168,168,166],2.23,[170,173,170,403,168,166,168,168,166],0.93,[173,173,170,405,168,166,168,168,166],1.47,[189,173,170,199,168,166,168,168,166],[283,173,170,408,168,166,168,168,166],0.43,[286,173,170,410,168,166,168,168,166],0.35,[289,173,170,412,168,166,168,168,166],0.17,[166,173,173,414,168,166,168,168,166],4.71,[170,173,173,416,168,166,168,168,166],1.71,[173,173,173,418,168,166,168,168,166],1.68,[189,173,173,420,168,166,168,168,166],1.2,[283,173,173,422,168,166,168,168,166],0.34,[286,173,173,424,168,166,168,168,166],0.28,[289,173,173,397,168,166,168,168,166],[166,173,189,427,168,166,168,168,166],3.69,[170,173,189,429,168,166,168,168,166],1.34,[173,173,189,431,168,166,168,168,166],1.33,[189,173,189,433,168,166,168,168,166],0.92,[283,173,189,435,168,166,168,168,166],0.47,[286,173,189,437,168,166,168,168,166],0.41,[289,173,189,439,168,166,168,168,166],0.32,[],[442],"Table 2",[],[],[446],"ICL-NUIM living room with simulated depth noise; ground-truth trajectories used and loop-closure handling disabled for all methods; reconstructions aligned to the ground-truth model with point-to-plane ICP; evaluation threshold 1 cm; 'smoothed' = same bilateral filter as SurfelMeshing preprocessing",{"slug":448,"group":449,"sourceId":5,"sourceLabel":6,"table":450,"selfRows":451,"metrics":452,"seqs":465,"entrants":475,"cells":480,"outcomes":542,"locators":543,"hardware":544,"wordings":545,"notes":546},"infinitam2015-table-5","infinitam2015:Table 5","Table 5",16,[453,456,459,462],{"label":454,"unit":455,"statistic":142,"alignment":143},"Error (m), mean","m",{"label":457,"unit":455,"statistic":458,"alignment":143},"Error (m), median","median",{"label":460,"unit":455,"statistic":461,"alignment":143},"Error (m), std","std",{"label":463,"unit":455,"statistic":464,"alignment":143},"Error (m), max","max",[466,469,471,473],{"dataset":248,"sequence":467,"environment":468},"kt0 (living room)","synthetic living room",{"dataset":248,"sequence":470,"environment":468},"kt1 (living room)",{"dataset":248,"sequence":472,"environment":468},"kt2 (living room)",{"dataset":248,"sequence":474,"environment":468},"kt3 (living room)",[476,478],{"name":477,"methodId":79,"linkable":85,"proposed":85,"self":85},"Handa best (best value reported in the ICL-NUIM benchmark paper)",{"name":479,"methodId":5,"linkable":89,"proposed":89,"self":89},"ITM (InfiniTAM)",[481,483,485,487,489,491,493,495,497,499,500,501,503,505,507,509,511,513,515,517,518,520,522,524,526,528,530,532,534,536,538,540],[166,166,166,482,168,166,168,168,166],0.0114,[170,166,166,484,168,166,168,168,166],0.006,[166,166,170,486,168,166,168,168,166],0.008,[170,166,170,488,168,166,168,168,166],0.0057,[166,166,173,490,168,166,168,168,166],0.0085,[170,166,173,492,168,166,168,168,166],0.0048,[166,166,189,494,168,166,168,168,166],0.1503,[170,166,189,496,168,166,168,168,166],0.0585,[166,170,166,498,168,166,168,168,166],0.0084,[170,170,166,484,168,166,168,168,166],[166,170,170,492,168,166,168,168,166],[170,170,170,502,168,166,168,168,166],0.0049,[166,170,173,504,168,166,168,168,166],0.0071,[170,170,173,506,168,166,168,168,166],0.0038,[166,170,189,508,168,166,168,168,166],0.0124,[170,170,189,510,168,166,168,168,166],0.0535,[166,173,166,512,168,166,168,168,166],0.0171,[170,173,166,514,168,166,168,168,166],0.0046,[166,173,170,516,168,166,168,168,166],0.0286,[170,173,170,506,168,166,168,168,166],[166,173,173,519,168,166,168,168,166],0.0136,[170,173,173,521,168,166,168,168,166],0.0037,[166,173,189,523,168,166,168,168,166],0.2745,[170,173,189,525,168,166,168,168,166],0.0415,[166,189,166,527,168,166,168,168,166],1.0377,[170,189,166,529,168,166,168,168,166],0.0405,[166,189,170,531,168,166,168,168,166],1.0911,[170,189,170,533,168,166,168,168,166],0.0261,[166,189,173,535,168,166,168,168,166],1.0798,[170,189,173,537,168,166,168,168,166],0.0429,[166,189,189,539,168,166,168,168,166],1.0499,[170,189,189,541,168,166,168,168,166],0.3522,[],[450],[],[],[547],"Reconstruction error of the living room models against the ICL-NUIM ground-truth surface (m); statistics as labelled in the table",{"slug":549,"group":550,"sourceId":551,"sourceLabel":552,"table":553,"selfRows":554,"metrics":555,"seqs":559,"entrants":582,"cells":589,"outcomes":646,"locators":648,"hardware":649,"wordings":650,"notes":651},"supereight2018-table-i","supereight2018:Table I","supereight2018","Vespa et al., 2018","Table I",10,[556],{"label":557,"unit":455,"statistic":558,"alignment":143},"ATE (m)","RMSE",[560,563,565,567,569,572,574,576,578,580],{"dataset":248,"sequence":561,"environment":562},"ICL_LR_0","synthetic living room (ICL) and real indoor office scenes (TUM)",{"dataset":248,"sequence":564,"environment":562},"ICL_LR_1",{"dataset":248,"sequence":566,"environment":562},"ICL_LR_2",{"dataset":248,"sequence":568,"environment":562},"ICL_LR_3",{"dataset":570,"sequence":571,"environment":562},"TUM RGB-D","TUM_fr1_xyz",{"dataset":570,"sequence":573,"environment":562},"TUM_fr1_floor",{"dataset":570,"sequence":575,"environment":562},"TUM_fr1_plant",{"dataset":570,"sequence":577,"environment":562},"TUM_fr1_desk",{"dataset":570,"sequence":579,"environment":562},"TUM_fr2_desk",{"dataset":570,"sequence":581,"environment":562},"TUM_fr3_office",[583,585,587],{"name":584,"methodId":551,"linkable":89,"proposed":89,"self":85},"TSDF (octree TSDF fusion, ours)",{"name":586,"methodId":551,"linkable":89,"proposed":89,"self":85},"OFusion (octree occupancy fusion, ours)",{"name":588,"methodId":5,"linkable":89,"proposed":85,"self":89},"InfiniTAM [16] (default depth-only tracker)",[590,592,594,596,598,600,602,604,606,608,610,612,614,616,618,620,621,622,623,624,625,626,629,631,633,635,637,639,642,644],[166,166,166,591,168,166,168,168,166],0.0113,[170,166,166,593,168,166,168,168,166],0.0305,[173,166,166,595,168,166,168,168,166],0.3052,[166,166,170,597,168,166,168,168,166],0.0117,[170,166,170,599,168,166,168,168,166],0.0207,[173,166,170,601,168,166,168,168,166],0.0214,[166,166,173,603,168,166,168,168,166],0.004,[170,166,173,605,168,166,168,168,166],0.005,[173,166,173,607,168,166,168,168,166],0.1725,[166,166,189,609,168,166,168,168,166],0.7582,[170,166,189,611,168,166,168,168,166],0.0786,[173,166,189,613,168,166,168,168,166],0.4858,[166,166,283,615,168,166,168,168,166],0.0295,[170,166,283,617,168,166,168,168,166],0.0293,[173,166,283,619,168,166,168,168,166],0.0273,[166,166,286,79,166,166,168,168,166],[170,166,286,79,166,166,168,168,166],[173,166,286,79,166,166,168,168,166],[166,166,289,79,166,166,168,168,166],[170,166,289,79,166,166,168,168,166],[173,166,289,79,166,166,168,168,166],[166,166,627,628,168,166,168,168,166],7,0.103,[170,166,627,630,168,166,168,168,166],0.0995,[173,166,627,632,168,166,168,168,166],0.0647,[166,166,131,634,168,166,168,168,166],0.0641,[170,166,131,636,168,166,168,168,166],0.0902,[173,166,131,638,168,166,168,168,166],0.0598,[166,166,640,641,168,166,168,168,166],9,0.0686,[170,166,640,643,168,166,168,168,166],0.0604,[173,166,640,645,168,166,168,168,166],0.0996,[647],"tracking failure",[553],[],[],[652],"ATE RMSE (Euclidean distance between ground-truth and estimated positions) on ICL-NUIM living room and TUM RGB-D; 1 cm finest voxels, depth-only tracking, same parameters throughout; 'x' = tracking failure",[654,659,664,670],{"group":655,"slug":656,"sourceLabel":6,"table":657,"selfRows":131,"datasets":658},"infinitam2015:Table 3","infinitam2015-table-3","Table 3",[248],{"group":660,"slug":661,"sourceLabel":6,"table":662,"selfRows":131,"datasets":663},"infinitam2015:Table 4","infinitam2015-table-4","Table 4",[248],{"group":665,"slug":666,"sourceLabel":6,"table":667,"selfRows":289,"datasets":668},"infinitam2015:Text Sec.7.4","infinitam2015-text-sec-7-4","Text Sec.7.4",[669],"authors' swivel-chair test",{"group":671,"slug":672,"sourceLabel":234,"table":136,"selfRows":289,"datasets":673},"surfelmeshing2020:Table 1","surfelmeshing2020-table-1",[570],1790510658343]