[{"data":1,"prerenderedAt":690},["ShallowReactive",2],{"method-coslam2023":3},{"method":4,"reference":52,"equipment":72,"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":22,"limitations":26,"sensors":32,"platform":34,"estimator":35,"association":36,"timeModel":37,"deskew":38,"loopClosure":39,"globalOptimization":40,"mapRepresentation":41,"prior":42,"outputGeometry":43,"compute":44,"codeUrl":45,"codeLicense":46,"relatedVersions":47},"coslam2023","Wang et al., 2023a","Co-SLAM","Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAM",2023,"recent","C09","odometry_with_local_mapping","Co-SLAM 結合多解析度雜湊網格（hash grid）與 one-blob 座標編碼，兼顧收斂速度與表面連續補洞，並以隨機取樣所有關鍵影格光線進行全域光束調整。作者特別指出評估前的網格裁切（mesh culling）策略會改變重建指標，所有方法在其新裁切策略下指標都變好，顯示神經 SLAM 幾何數字高度依賴評估協定。","Hash-grid plus coordinate encoding with global BA over all keyframes; authors show mesh-culling choices materially change reconstruction metrics.","full_text_reviewed","peer_reviewed_published","main_body","論文未涉及營建場域；定量評估使用 Replica、NeuralRGBD 合成資料、ScanNet 與 TUM RGB-D，另有以 RealSense D435i 自行拍攝的兩段室內房間序列與 NICE-SLAM 以 Azure Kinect 拍攝的公寓序列，但僅作定性比較。",[20,21],"simulation","public_benchmark",[23,24,25],"Real-time (17 FPS on Replica) with high-fidelity reconstruction and plausible hole filling (abstract; Sec. 4.1)","Replica Acc. 2.10 cm, Comp. 2.08 cm, completion ratio 93.44% versus NICE-SLAM 2.37 cm, 2.64 cm and 91.13% under the same culling (Table 1)","Global BA with rays from all keyframes gives ScanNet ATE 8.75 cm (std 0.33) versus 9.69 cm (std 1.38) for local BA over 10 keyframes (Table 6)",[27,28,29,30,31],"Relies on RGB-D input; sensitive to illumination changes and inaccurate depth (Sec. 5)","No loop closure (Sec. 5)","Reconstruction metrics depend on the mesh-culling strategy; all methods gain under the proposed culling (Sec. 4.1; Supp. 1.2)","Still less accurate than classic SLAM on TUM (ORB-SLAM2 1.6, 0.4 and 1.0 cm versus 2.7, 1.9 and 2.6 cm) (Table 4; Sec. 4.2)","Rigidly aligned ATE hides distortion: without alignment ScanNet ATE averages 18.01 cm (NICE-SLAM 23.97 cm) (Supp. 2.3, Supp. Table 3)",[33],"RGB-D",[],"gradient-based per-frame tracking (constant-speed initialization) + global bundle adjustment over rays sampled from all keyframes","direct colour and depth rendering losses + approximate SDF and smoothness losses","discrete poses","not_applicable","none (authors suggest incorporating loop closure as future work)","global bundle adjustment jointly optimizing the scene representation and all keyframe poses","multi-resolution hash grid + one-blob coordinate encoding with shallow MLPs predicting TSDF and colour","none (RGB-D measurements with known intrinsics)","mesh via marching cubes, evaluated after mesh culling","Intel Core i7-12700K (3.60 GHz) + NVIDIA RTX 3090 Ti; 17.4 FPS on Replica, 15.6 FPS on Synthetic RGB-D, 12.8 FPS on ScanNet and 13.3 FPS on TUM with default settings; about 6.4 to 6.7 FPS with doubled tracking iterations; 0.26 M to 1.6 M parameters (Tables 1 and 2)","https:\u002F\u002Fgithub.com\u002FHengyiWang\u002FCo-SLAM","Apache-2.0",[48],{"relation":49,"title":50,"doi_or_url":51},"preprint","arXiv:2304.14377","https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.14377",{"id":5,"kind":53,"shortName":7,"title":8,"authors":54,"year":9,"venue":58,"venueType":59,"publisher":60,"volumeIssuePages":61,"doi":62,"arxivId":63,"url":64,"firstPublicDate":65,"publicationStatus":16,"metadataStatus":66,"fulltextStatus":15,"era":10,"classicReason":38,"codeUrl":45,"cluster":11,"topics":67,"mdpi":68,"verification":69,"label":6,"fulltextRoute":70,"versionRead":71,"addedByCensus":68},"method",[55,56,57],"Hengyi Wang","Jingwen Wang","Lourdes Agapito","2023 IEEE\u002FCVF Conference on Computer Vision and Pattern Recognition (CVPR)","conference","IEEE","pp. 13293-13302","10.1109\u002Fcvpr52729.2023.01277","2304.14377","https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1109\u002Fcvpr52729.2023.01277","2023-04-27","metadata_verified",[11],false,"confirmed","arXiv","arXiv v1 (2304.14377v1, 27 Apr 2023; only version) including supplementary material; CVPR 2023 version of record (pp. 13293 to 13302) not compared",[73,80,83,90,96],{"category":74,"model":75,"canonical":75,"role":76,"dataset":77,"specs":78,"locator":79},"compute","Intel Core i7-12700K","compute for runtime",null,"3.60 GHz desktop CPU","Sec. 4.1; Sec. 4.3",{"category":74,"model":81,"canonical":81,"role":76,"dataset":77,"specs":82,"locator":79},"NVIDIA RTX 3090 Ti","desktop GPU used for all timings",{"category":84,"model":85,"canonical":86,"role":87,"dataset":77,"specs":88,"locator":89},"rgbd","RealSense D435i","Intel RealSense D435I","method input","depth quality described as slightly worse than Azure Kinect; two self-captured indoor room sequences, qualitative only","Supp. 2.2; Supp. Figs. 12 and 13",{"category":84,"model":91,"canonical":91,"role":92,"dataset":93,"specs":94,"locator":95},"Azure Kinect","dataset sensor","NICE-SLAM apartment sequence","apartment sequence captured by the NICE-SLAM authors; run with the ScanNet setting, qualitative only","Supp. 2.2; Supp. Fig. 11",{"category":97,"model":98,"canonical":98,"role":99,"dataset":100,"specs":101,"locator":102},"other","motion capture system (model not reported)","reference or ground truth","TUM RGB-D","provides TUM RGB-D ground-truth poses","Sec. 4.1",[],{"totalRows":105,"groupCount":106,"groups":107,"others":563},133,29,[108,257,351,413],{"slug":109,"group":110,"sourceId":5,"sourceLabel":6,"table":111,"selfRows":112,"metrics":113,"seqs":133,"entrants":140,"cells":155,"outcomes":249,"locators":250,"hardware":251,"wordings":254,"notes":255},"coslam2023-table-1","coslam2023:Table 1","Table 1",14,[114,118,120,122,125,128,131],{"label":115,"unit":116,"statistic":117,"alignment":38},"Depth L1 (cm)","cm","mean",{"label":119,"unit":116,"statistic":117,"alignment":38},"Acc. (cm)",{"label":121,"unit":116,"statistic":117,"alignment":38},"Comp. (cm)",{"label":123,"unit":124,"statistic":117,"alignment":38},"Comp. Ratio (%, \u003C 5 cm)","%",{"label":126,"unit":127,"statistic":117,"alignment":38},"FPS","Hz",{"label":129,"unit":130,"statistic":117,"alignment":38},"Tracking (ms) (1024, 10) as time(#pixel, #iter)","ms",{"label":132,"unit":130,"statistic":117,"alignment":38},"Mapping (ms) (2048, 10) as time(#pixel, #iter)",[134,138],{"dataset":135,"sequence":136,"environment":137},"Replica (8 synthetic scenes)","average over scenes","synthetic indoor",{"dataset":139,"sequence":136,"environment":137},"Synthetic RGB-D of NeuralRGBD (7 scenes, simulated depth noise)",[141,145,148,151,153],{"name":142,"methodId":143,"linkable":144,"proposed":68,"self":68},"TSDF-Fusion","curless1996volumetric",true,{"name":146,"methodId":147,"linkable":144,"proposed":68,"self":68},"iMAP","imap2021",{"name":149,"methodId":150,"linkable":144,"proposed":68,"self":68},"NICE-SLAM","niceslam2022",{"name":152,"methodId":5,"linkable":144,"proposed":144,"self":144},"Co-SLAM (Ours)",{"name":154,"methodId":147,"linkable":144,"proposed":68,"self":68},"iMAP*",[156,160,163,166,169,171,173,175,177,180,182,184,186,188,190,192,194,196,198,201,204,206,208,209,211,213,215,217,219,221,223,225,227,229,231,233,235,237,239,241,243,245,247],[157,157,157,158,159,157,159,159,157],0,6.36,-1,[157,161,157,162,159,157,159,159,157],1,1.62,[157,164,157,165,159,157,159,159,157],2,3.94,[157,167,157,168,159,157,159,159,157],3,83.93,[161,157,157,170,159,157,159,159,157],4.64,[161,161,157,172,159,157,159,159,157],3.62,[161,164,157,174,159,157,159,159,157],4.93,[161,167,157,176,159,157,159,159,157],80.51,[161,178,157,179,159,157,157,159,157],4,9.9,[164,157,157,181,159,157,159,159,157],1.9,[164,161,157,183,159,157,159,159,157],2.37,[164,164,157,185,159,157,159,159,157],2.64,[164,167,157,187,159,157,159,159,157],91.13,[164,178,157,189,159,157,157,159,157],0.91,[167,157,157,191,159,157,159,159,157],1.51,[167,161,157,193,159,157,159,159,157],2.1,[167,164,157,195,159,157,159,159,157],2.08,[167,167,157,197,159,157,159,159,157],93.44,[167,199,157,200,159,157,161,159,157],5,58,[167,202,157,203,159,157,161,159,157],6,98,[167,178,157,205,159,157,161,159,157],17.4,[157,157,161,207,159,157,159,159,157],10.87,[157,161,161,162,159,157,159,159,157],[157,164,161,210,159,157,159,159,157],5.16,[157,167,161,212,159,157,159,159,157],81.52,[178,157,161,214,159,157,159,159,157],43.91,[178,161,161,216,159,157,159,159,157],18.3,[178,164,161,218,159,157,159,159,157],26.41,[178,167,161,220,159,157,159,159,157],20.73,[178,178,161,222,159,157,157,159,157],0.34,[164,157,161,224,159,157,159,159,157],6.32,[164,161,161,226,159,157,159,159,157],5.96,[164,164,161,228,159,157,159,159,157],5.3,[164,167,161,230,159,157,159,159,157],77.46,[164,199,161,232,159,157,157,159,157],123,[164,178,161,234,159,157,157,159,157],1.31,[167,157,161,236,159,157,159,159,157],3.02,[167,161,161,238,159,157,159,159,157],2.95,[167,164,161,240,159,157,159,159,157],2.96,[167,167,161,242,159,157,159,159,157],86.88,[167,199,161,244,159,157,161,159,157],64,[167,202,161,246,159,157,161,159,157],104,[167,178,161,248,159,157,161,159,157],15.6,[],[111],[252,253],"not stated for baselines (the paper names the i7-12700K + RTX 3090ti desktop only for running Co-SLAM, Sec. 4.1)","Intel Core i7-12700K + NVIDIA RTX 3090 Ti",[],[256],"Averages over scenes; meshes culled with the authors' new culling strategy for all methods; TSDF-Fusion uses Co-SLAM poses; iMAP* is the NICE-SLAM re-implementation",{"slug":258,"group":259,"sourceId":5,"sourceLabel":6,"table":260,"selfRows":112,"metrics":261,"seqs":266,"entrants":283,"cells":289,"outcomes":345,"locators":346,"hardware":347,"wordings":348,"notes":349},"coslam2023-table-3","coslam2023:Table 3","Table 3",[262],{"label":263,"unit":116,"statistic":264,"alignment":265},"ATE RMSE (cm), average of 5 runs","RMSE","SE3",[267,271,273,275,277,279,281],{"dataset":268,"sequence":269,"environment":270},"ScanNet","scene0000","real indoor rooms",{"dataset":268,"sequence":272,"environment":270},"scene0059",{"dataset":268,"sequence":274,"environment":270},"scene0106",{"dataset":268,"sequence":276,"environment":270},"scene0169",{"dataset":268,"sequence":278,"environment":270},"scene0181",{"dataset":268,"sequence":280,"environment":270},"scene0207",{"dataset":268,"sequence":282,"environment":270},"average",[284,285,286,288],{"name":154,"methodId":147,"linkable":144,"proposed":68,"self":68},{"name":149,"methodId":150,"linkable":144,"proposed":68,"self":68},{"name":287,"methodId":5,"linkable":144,"proposed":144,"self":144},"Co-SLAM (Ours-dagger)",{"name":152,"methodId":5,"linkable":144,"proposed":144,"self":144},[290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338,340,342,343],[157,157,157,291,159,157,159,159,157],55.95,[157,157,161,293,159,157,159,159,157],32.06,[157,157,164,295,159,157,159,159,157],17.5,[157,157,167,297,159,157,159,159,157],70.51,[157,157,178,299,159,157,159,159,157],32.1,[157,157,199,301,159,157,159,159,157],11.91,[157,157,202,303,159,157,159,159,157],36.67,[161,157,157,305,159,157,159,159,157],8.64,[161,157,161,307,159,157,159,159,157],12.25,[161,157,164,309,159,157,159,159,157],8.09,[161,157,167,311,159,157,159,159,157],10.28,[161,157,178,313,159,157,159,159,157],12.93,[161,157,199,315,159,157,159,159,157],5.59,[161,157,202,317,159,157,159,159,157],9.63,[164,157,157,319,159,157,159,159,157],7.13,[164,157,161,321,159,157,159,159,157],11.14,[164,157,164,323,159,157,159,159,157],9.36,[164,157,167,325,159,157,159,159,157],5.9,[164,157,178,327,159,157,159,159,157],11.81,[164,157,199,329,159,157,159,159,157],7.14,[164,157,202,331,159,157,159,159,157],8.75,[167,157,157,333,159,157,159,159,157],7.18,[167,157,161,335,159,157,159,159,157],12.29,[167,157,164,337,159,157,159,159,157],9.57,[167,157,167,339,159,157,159,159,157],6.62,[167,157,178,341,159,157,159,159,157],13.43,[167,157,199,319,159,157,159,159,157],[167,157,202,344,159,157,159,159,157],9.37,[],[260],[],[],[350],"ATE RMSE averaged over 5 runs; ground truth from BundleFusion; rigid alignment of trajectory (Supp. Sec. 2.3)",{"slug":352,"group":353,"sourceId":5,"sourceLabel":6,"table":354,"selfRows":355,"metrics":356,"seqs":367,"entrants":373,"cells":378,"outcomes":407,"locators":408,"hardware":409,"wordings":410,"notes":411},"coslam2023-table-2","coslam2023:Table 2","Table 2",12,[357,358,361,363,365],{"label":126,"unit":127,"statistic":117,"alignment":38},{"label":359,"unit":360,"statistic":117,"alignment":38},"Track. (ms\u002Fiter) x 20 iter","ms per iteration",{"label":362,"unit":360,"statistic":117,"alignment":38},"Map. (ms\u002Fiter) x 10 iter",{"label":364,"unit":360,"statistic":117,"alignment":38},"Track. (ms\u002Fiter) x 10 iter",{"label":366,"unit":360,"statistic":117,"alignment":38},"Map. (ms\u002Fiter) x 20 iter",[368,371],{"dataset":268,"sequence":369,"environment":370},"scenes used in paper","real-world indoor sequences (ScanNet)",{"dataset":100,"sequence":369,"environment":372},"real-world indoor sequences (TUM RGB-D)",[374,375,376,377],{"name":154,"methodId":147,"linkable":144,"proposed":68,"self":68},{"name":149,"methodId":150,"linkable":144,"proposed":68,"self":68},{"name":287,"methodId":5,"linkable":144,"proposed":144,"self":144},{"name":152,"methodId":5,"linkable":144,"proposed":144,"self":144},[379,381,383,385,387,389,390,391,393,395,397,399,401,403,404,405],[157,157,157,380,159,157,157,159,157],0.37,[161,157,157,382,159,157,157,159,157],0.68,[164,161,157,384,159,157,161,159,157],7.8,[164,164,157,386,159,157,161,159,157],20.2,[164,157,157,388,159,157,161,159,157],6.4,[167,167,157,384,159,157,161,159,157],[167,164,157,386,159,157,161,159,157],[167,157,157,392,159,157,161,159,157],12.8,[157,157,161,394,159,157,157,159,157],0.07,[161,157,161,396,159,157,157,159,157],0.08,[164,161,161,398,159,157,161,159,157],7.5,[164,178,161,400,159,157,161,159,157],19,[164,157,161,402,159,157,161,159,157],6.7,[167,167,161,398,159,157,161,159,157],[167,178,161,400,159,157,161,159,157],[167,157,161,406,159,157,161,159,157],13.3,[],[354],[252,253],[],[412],"Run-time as ms\u002Fiter x #iter; NICE-SLAM and iMAP* map every frame on TUM, otherwise mapping every 5 frames; Ours-dagger uses twice the tracking iterations",{"slug":414,"group":415,"sourceId":416,"sourceLabel":417,"table":111,"selfRows":418,"metrics":419,"seqs":423,"entrants":444,"cells":459,"outcomes":557,"locators":558,"hardware":559,"wordings":560,"notes":561},"gsslam2024-table-1","gsslam2024:Table 1","gsslam2024","Yan et al., 2024",9,[420],{"label":421,"unit":116,"statistic":264,"alignment":422},"ATE RMSE [cm]","not_reported",[424,428,430,432,434,436,438,440,442],{"dataset":425,"sequence":426,"environment":427},"Replica","Rm0","synthetic indoor scenes",{"dataset":425,"sequence":429,"environment":427},"Rm1",{"dataset":425,"sequence":431,"environment":427},"Rm2",{"dataset":425,"sequence":433,"environment":427},"Off0",{"dataset":425,"sequence":435,"environment":427},"Off1",{"dataset":425,"sequence":437,"environment":427},"Off2",{"dataset":425,"sequence":439,"environment":427},"Off3",{"dataset":425,"sequence":441,"environment":427},"Off4",{"dataset":425,"sequence":443,"environment":427},"average of 8 scenes",[445,448,450,452,455,457],{"name":446,"methodId":447,"linkable":144,"proposed":68,"self":68},"Point-SLAM [ 27 ]","pointslam2023",{"name":449,"methodId":150,"linkable":144,"proposed":68,"self":68},"NICE-SLAM [ 55 ]",{"name":451,"methodId":77,"linkable":68,"proposed":68,"self":68},"Vox-Fusion ∗ [ 48 ]",{"name":453,"methodId":454,"linkable":144,"proposed":68,"self":68},"ESLAM [ 11 ]","eslam2023",{"name":456,"methodId":5,"linkable":144,"proposed":68,"self":144},"CoSLAM [ 41 ]",{"name":458,"methodId":77,"linkable":68,"proposed":144,"self":68},"Ours",[460,462,464,466,468,470,472,474,477,480,482,483,485,487,488,490,492,494,495,497,499,501,503,505,507,509,511,513,515,517,519,521,522,524,525,527,528,529,531,533,535,536,538,540,541,542,544,546,548,549,551,552,554,555],[157,157,157,461,159,157,159,159,157],0.56,[157,157,161,463,159,157,159,159,157],0.47,[157,157,164,465,159,157,159,159,157],0.3,[157,157,167,467,159,157,159,159,157],0.35,[157,157,178,469,159,157,159,159,157],0.62,[157,157,199,471,159,157,159,159,157],0.55,[157,157,202,473,159,157,159,159,157],0.72,[157,157,475,476,159,157,159,159,157],7,0.73,[157,157,478,479,159,157,159,159,157],8,0.54,[161,157,157,481,159,157,159,159,157],0.97,[161,157,161,234,159,157,159,159,157],[161,157,164,484,159,157,159,159,157],1.07,[161,157,167,486,159,157,159,159,157],0.88,[161,157,178,161,159,157,159,159,157],[161,157,199,489,159,157,159,159,157],1.06,[161,157,202,491,159,157,159,159,157],1.1,[161,157,475,493,159,157,159,159,157],1.13,[161,157,478,489,159,157,159,159,157],[164,157,157,496,159,157,159,159,157],1.37,[164,157,161,498,159,157,159,159,157],4.7,[164,157,164,500,159,157,159,159,157],1.47,[164,157,167,502,159,157,159,159,157],8.48,[164,157,178,504,159,157,159,159,157],2.04,[164,157,199,506,159,157,159,159,157],2.58,[164,157,202,508,159,157,159,159,157],1.11,[164,157,475,510,159,157,159,159,157],2.94,[164,157,478,512,159,157,159,159,157],3.09,[167,157,157,514,159,157,159,159,157],0.71,[167,157,161,516,159,157,159,159,157],0.7,[167,157,164,518,159,157,159,159,157],0.52,[167,157,167,520,159,157,159,159,157],0.57,[167,157,178,471,159,157,159,159,157],[167,157,199,523,159,157,159,159,157],0.58,[167,157,202,473,159,157,159,159,157],[167,157,475,526,159,157,159,159,157],0.63,[167,157,478,526,159,157,159,159,157],[178,157,157,516,159,157,159,159,157],[178,157,161,530,159,157,159,159,157],0.95,[178,157,164,532,159,157,159,159,157],1.35,[178,157,167,534,159,157,159,159,157],0.59,[178,157,178,471,159,157,159,159,157],[178,157,199,537,159,157,159,159,157],2.03,[178,157,202,539,159,157,159,159,157],1.56,[178,157,475,473,159,157,159,159,157],[178,157,478,161,159,157,159,159,157],[199,157,157,543,159,157,159,159,157],0.48,[199,157,161,545,159,157,159,159,157],0.53,[199,157,164,547,159,157,159,159,157],0.33,[199,157,167,518,159,157,159,159,157],[199,157,178,550,159,157,159,159,157],0.41,[199,157,199,534,159,157,159,159,157],[199,157,202,553,159,157,159,159,157],0.46,[199,157,475,516,159,157,159,159,157],[199,157,478,556,159,157,159,159,157],0.5,[],[111],[],[],[562],"Replica ATE RMSE, 8 scenes; * = reproduced with official code; methods in upper part run below 5 FPS",[564,571,576,581,587,592,596,600,605,610,615,619,624,629,635,639,644,648,653,660,664,670,675,679,685],{"group":565,"slug":566,"sourceLabel":567,"table":568,"selfRows":418,"datasets":569},"yan2026_underground3dgsslam:Table III","yan2026-underground3dgsslam-table-iii","Yan et al., 2026b","Table III",[570],"Underground_RGB-D (authors' field test dataset)",{"group":572,"slug":573,"sourceLabel":6,"table":574,"selfRows":475,"datasets":575},"coslam2023:Supp. Table 3","coslam2023-supp-table-3","Supp. Table 3",[268],{"group":577,"slug":578,"sourceLabel":6,"table":579,"selfRows":202,"datasets":580},"coslam2023:Table 4","coslam2023-table-4","Table 4",[100],{"group":582,"slug":583,"sourceLabel":584,"table":111,"selfRows":202,"datasets":585},"rtgslam2024:Table 1","rtgslam2024-table-1","Peng et al., 2024",[586,425],"Azure (self-scanned)",{"group":588,"slug":589,"sourceLabel":6,"table":590,"selfRows":178,"datasets":591},"coslam2023:Supp. Table 1","coslam2023-supp-table-1","Supp. Table 1",[135],{"group":593,"slug":594,"sourceLabel":417,"table":354,"selfRows":178,"datasets":595},"gsslam2024:Table 2","gsslam2024-table-2",[100],{"group":597,"slug":598,"sourceLabel":417,"table":260,"selfRows":178,"datasets":599},"gsslam2024:Table 3","gsslam2024-table-3",[425],{"group":601,"slug":602,"sourceLabel":603,"table":111,"selfRows":178,"datasets":604},"monogs2024:Table 1","monogs2024-table-1","Matsuki et al., 2024",[100],{"group":606,"slug":607,"sourceLabel":608,"table":111,"selfRows":178,"datasets":609},"photoslam2024:Table 1","photoslam2024-table-1","Huang et al., 2024c",[425],{"group":611,"slug":612,"sourceLabel":584,"table":613,"selfRows":178,"datasets":614},"rtgslam2024:Supp. Table 5","rtgslam2024-supp-table-5","Supp. Table 5",[425],{"group":616,"slug":617,"sourceLabel":584,"table":354,"selfRows":178,"datasets":618},"rtgslam2024:Table 2","rtgslam2024-table-2",[100],{"group":620,"slug":621,"sourceLabel":584,"table":260,"selfRows":178,"datasets":622},"rtgslam2024:Table 3","rtgslam2024-table-3",[623],"ScanNet++",{"group":625,"slug":626,"sourceLabel":627,"table":111,"selfRows":167,"datasets":628},"deng2026_mcgs_slam:Table 1","deng2026-mcgs-slam-table-1","Deng & Gan, 2026",[100],{"group":630,"slug":631,"sourceLabel":632,"table":354,"selfRows":167,"datasets":633},"loopyslam2024:Table 2","loopyslam2024-table-2","Liso et al., 2024",[634],"TUM-RGBD",{"group":636,"slug":637,"sourceLabel":608,"table":354,"selfRows":167,"datasets":638},"photoslam2024:Table 2","photoslam2024-table-2",[100],{"group":640,"slug":641,"sourceLabel":567,"table":642,"selfRows":167,"datasets":643},"yan2026_underground3dgsslam:Table VI","yan2026-underground3dgsslam-table-vi","Table VI",[100],{"group":645,"slug":646,"sourceLabel":417,"table":579,"selfRows":164,"datasets":647},"gsslam2024:Table 4","gsslam2024-table-4",[425],{"group":649,"slug":650,"sourceLabel":584,"table":651,"selfRows":164,"datasets":652},"rtgslam2024:Supp. Table 7","rtgslam2024-supp-table-7","Supp. Table 7",[100],{"group":654,"slug":655,"sourceLabel":656,"table":657,"selfRows":164,"datasets":658},"tosi2026survey:Table XI","tosi2026survey-table-xi","Tosi et al., 2026","Table XI",[659,425],"KITTI",{"group":661,"slug":662,"sourceLabel":627,"table":354,"selfRows":161,"datasets":663},"deng2026_mcgs_slam:Table 2","deng2026-mcgs-slam-table-2",[100],{"group":665,"slug":666,"sourceLabel":667,"table":668,"selfRows":161,"datasets":669},"hislam2_2025:Table II","hislam2-2025-table-ii","Zhang et al., 2025","Table II",[268],{"group":671,"slug":672,"sourceLabel":632,"table":673,"selfRows":161,"datasets":674},"loopyslam2024:Table 11","loopyslam2024-table-11","Table 11",[425],{"group":676,"slug":677,"sourceLabel":632,"table":260,"selfRows":161,"datasets":678},"loopyslam2024:Table 3","loopyslam2024-table-3",[268],{"group":680,"slug":681,"sourceLabel":682,"table":683,"selfRows":161,"datasets":684},"pinslam2024:Table IX","pinslam2024-table-ix","Pan et al., 2024","Table IX",[425],{"group":686,"slug":687,"sourceLabel":584,"table":688,"selfRows":161,"datasets":689},"rtgslam2024:Supp. Table 6","rtgslam2024-supp-table-6","Supp. Table 6",[425],1790510664704]