[{"data":1,"prerenderedAt":1045},["ShallowReactive",2],{"method-kerbl2023_3dgs":3},{"method":4,"reference":50,"equipment":71,"figures":85,"results":124},{"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":21,"limitations":25,"sensors":32,"platform":34,"estimator":35,"association":36,"timeModel":37,"deskew":37,"loopClosure":38,"globalOptimization":38,"mapRepresentation":39,"prior":40,"outputGeometry":41,"compute":42,"codeUrl":43,"codeLicense":44,"relatedVersions":45},"kerbl2023_3dgs","Kerbl et al., 2023","3DGS","3D Gaussian Splatting for Real-Time Radiance Field Rendering",2023,"recent","C09","map_representation_or_reconstruction","3DGS 以具各向異性共變異的三維高斯基元表示場景，並以可微分的分塊光柵化（tile-based rasterization）直接由影像誤差最佳化其位置、形狀、不透明度與球諧顏色。初始化依賴 SfM 相機與稀疏點雲。其目標是即時新視角渲染，而非量測等級的表面幾何。","Optimizes explicit anisotropic 3D Gaussians initialized from SfM points with a fast differentiable rasterizer for real-time novel-view synthesis.","full_text_reviewed","peer_reviewed_published","background","原論文未涉及營建；AEC 中 3DGS 與 LiDAR 的比較見 [yu2025_3dgs_lidar_heritage]。",[20],"public_benchmark",[22,23,24],"Real-time radiance-field rendering with high visual quality (abstract)","On Mip-NeRF360, PSNR 27.21 at 30K iterations vs 27.69 reported for Mip-NeRF360, with 41 min vs 48 h training and 134 FPS vs 0.06 FPS (Table 1)","At 7K iterations quality is comparable to InstantNGP and Plenoxels with similar training time (Sec. 7.2, Table 1)",[26,27,28,29,30,31],"Artifacts in poorly observed regions; elongated or 'splotchy' Gaussians; popping artifacts (Sec. 7.4)","No regularization applied to the optimization (Sec. 7.4)","Follow-up work states 3DGS fails to accurately represent surfaces because 3D Gaussians are multi-view inconsistent ([huang2024_2dgs] abstract)","Memory consumption is significantly higher than NeRF-based solutions: peak training memory can exceed 20 GB on large scenes and stored models reach hundreds of MB (Sec. 7.4, Table 1)","Very large scenes such as urban datasets may need a reduced position learning rate (Sec. 7.4)","Views with little overlap with training views show artifacts (Fig. 12)",[33],"monocular camera (multi-view images)",[],"not_applicable (gradient-based optimization of Gaussian parameters with adaptive density control; cameras from SfM)","direct photometric loss via differentiable tile-based rasterization","not_applicable","none","explicit anisotropic 3D Gaussians (position, covariance, opacity, spherical-harmonic colour)","SfM-calibrated cameras (SfM of Schönberger and Frahm 2016; the paper does not name the software) and the SfM sparse point cloud for initialization; random initialization is used for synthetic Blender scenes and degrades real scenes mainly in the background","rendered images and a set of 1 to 5 million anisotropic Gaussians; no surface or mesh extraction; the authors list mesh reconstruction from the Gaussians as future work","PyTorch with custom CUDA rasterization kernels (NVIDIA CUB radix sort); all reported results on an A6000 GPU (Mip-NeRF360 baseline trained on a 4-GPU A100 node); real-time rendering of at least 30 fps at 1080p, 134 to 154 FPS at 30K iterations on the three real datasets; training 26m54s to 41m33s at 30K iterations; peak training memory can exceed 20 GB on large scenes","https:\u002F\u002Fgithub.com\u002Fgraphdeco-inria\u002Fgaussian-splatting","Gaussian-Splatting License (custom Inria\u002FMPII licence: research use, non-commercial only; commercial use prohibited without licensor consent; LICENSE.md read)",[46],{"relation":47,"title":48,"doi_or_url":49},"preprint","arXiv:2308.04079 (posted after the TOG online date)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.04079",{"id":5,"kind":51,"shortName":7,"title":8,"authors":52,"year":9,"venue":57,"venueType":58,"publisher":59,"volumeIssuePages":60,"doi":61,"arxivId":62,"url":63,"firstPublicDate":64,"publicationStatus":16,"metadataStatus":65,"fulltextStatus":15,"era":10,"classicReason":37,"codeUrl":43,"cluster":11,"topics":66,"mdpi":67,"verification":68,"label":6,"fulltextRoute":69,"versionRead":70,"addedByCensus":67},"method",[53,54,55,56],"Bernhard Kerbl","Georgios Kopanas","Thomas Leimkuehler","George Drettakis","ACM Transactions on Graphics","journal","ACM","42(4), 1-14","10.1145\u002F3592433","2308.04079","https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1145\u002F3592433","2023-07-26","metadata_verified",[11],false,"corrected","arXiv","arXiv v1 HTML (2023-08-08), which carries the ACM TOG 42(4) camera-ready metadata; ACM version of record not opened",[72,80],{"category":73,"model":74,"canonical":75,"role":76,"dataset":77,"specs":78,"locator":79},"compute","A6000 GPU (as written)","A6000 GPU","compute for runtime",null,"used for all reported results except the Mip-NeRF360 baseline","Sec. 7.2",{"category":73,"model":81,"canonical":82,"role":76,"dataset":77,"specs":83,"locator":84},"4-GPU A100 node (as written)","4-GPU A100 node","used to train the Mip-NeRF360 baseline for 12 h","Sec. 7.2, footnote 2",[86,99,108,116],{"refId":5,"refLabel":6,"fig":87,"whatZh":88,"license":89,"licenseUrl":90,"sourceUrl":91,"src":92,"width":93,"height":94,"thumb":95,"thumbWidth":96,"thumbHeight":97,"modified":98},"Fig. 1","3DGS 即時渲染品質與訓練時間，與 Mip-NeRF360、InstantNGP、Plenoxels 比較","CC BY 4.0 (arXiv v1)","http:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002F","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2308.04079v1\u002Fteaser_02.png","\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-1.webp",842,193,"\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-1.thumb.webp",480,110,"converted to WebP",{"refId":5,"refLabel":6,"fig":100,"whatZh":101,"license":89,"licenseUrl":90,"sourceUrl":102,"src":103,"width":104,"height":105,"thumb":106,"thumbWidth":96,"thumbHeight":107,"modified":98},"Fig. 2","方法總覽：由 SfM 稀疏點建立三維高斯，最佳化並自適應控制密度，再以分塊光柵化渲染","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2308.04079v1\u002Foverview_01.png","\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-2.webp",975,180,"\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-2.thumb.webp",89,{"refId":5,"refLabel":6,"fig":109,"whatZh":110,"license":89,"licenseUrl":90,"sourceUrl":111,"src":112,"width":113,"height":114,"thumb":115,"thumbWidth":113,"thumbHeight":114,"modified":98},"Fig. 3","將最佳化後的三維高斯縮小 60% 以顯示其各向異性形狀","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2308.04079v1\u002Ffigures\u002Fanisotropic\u002Freal2.png","\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-3.webp",326,131,"\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-3.thumb.webp",{"refId":5,"refLabel":6,"fig":117,"whatZh":118,"license":89,"licenseUrl":90,"sourceUrl":119,"src":120,"width":121,"height":122,"thumb":123,"thumbWidth":121,"thumbHeight":122,"modified":98},"Fig. 4","自適應密度控制：重建不足時複製高斯，過度重建時分裂高斯","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2308.04079v1\u002Fdensity_control_01.png","\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-4.webp",467,275,"\u002Ffigure-files\u002Fkerbl2023_3dgs\u002Ffig-4.thumb.webp",{"totalRows":125,"groupCount":126,"groups":127,"others":991},130,12,[128,379,552,832],{"slug":129,"group":130,"sourceId":5,"sourceLabel":6,"table":131,"selfRows":132,"metrics":133,"seqs":161,"entrants":172,"cells":186,"outcomes":372,"locators":373,"hardware":374,"wordings":376,"notes":377},"kerbl2023-3dgs-table-1","kerbl2023_3dgs:Table 1","Table 1",36,[134,137,140,142,145,148,151,153,155,157,159],{"label":135,"unit":38,"statistic":136,"alignment":38},"SSIM","mean",{"label":138,"unit":139,"statistic":136,"alignment":38},"PSNR","dB",{"label":141,"unit":38,"statistic":136,"alignment":38},"LPIPS",{"label":143,"unit":144,"statistic":136,"alignment":38},"FPS (rendering)","fps",{"label":146,"unit":147,"statistic":136,"alignment":38},"Mem (memory to store optimized parameters)","MB",{"label":149,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '6m25s')","min",{"label":152,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '6m55s')",{"label":154,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '4m35s')",{"label":156,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '41m33s')",{"label":158,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '26m54s')",{"label":160,"unit":150,"statistic":136,"alignment":38},"Train (training time, written '36m2s')",[162,166,169],{"dataset":163,"sequence":164,"environment":165},"Mip-NeRF360","dataset average","full Mip-NeRF360 real-scene set: bicycle, flowers, garden, stump, treehill, room, counter, kitchen, bonsai (Appendix D; the paper states only that its 13 real test scenes overall cover bounded indoor and unbounded outdoor environments)",{"dataset":167,"sequence":164,"environment":168},"Tanks&Temples","real scenes Truck and Train",{"dataset":170,"sequence":164,"environment":171},"Deep Blending","real scenes Playroom and DrJohnson",[173,175,177,179,181,184],{"name":174,"methodId":77,"linkable":67,"proposed":67,"self":67},"Plenoxels",{"name":176,"methodId":77,"linkable":67,"proposed":67,"self":67},"INGP-Base",{"name":178,"methodId":77,"linkable":67,"proposed":67,"self":67},"INGP-Big",{"name":180,"methodId":77,"linkable":67,"proposed":67,"self":67},"M-NeRF360",{"name":182,"methodId":5,"linkable":183,"proposed":183,"self":183},"Ours-7K",true,{"name":185,"methodId":5,"linkable":183,"proposed":183,"self":183},"Ours-30K",[187,191,194,197,200,202,204,206,208,210,212,214,216,218,220,222,224,226,228,230,232,234,235,237,239,241,243,244,246,248,250,252,254,256,258,260,262,263,265,267,269,271,272,274,276,278,280,282,284,285,287,289,291,293,294,296,298,300,303,305,307,309,311,313,316,318,320,322,324,326,329,331,333,335,337,339,342,344,346,348,350,352,355,357,359,361,363,365,368,370],[188,188,188,189,190,188,190,190,188],0,0.626,-1,[188,192,188,193,190,188,190,190,188],1,23.08,[188,195,188,196,190,188,190,190,188],2,0.463,[188,198,188,199,190,188,188,190,188],3,6.79,[188,188,192,201,190,188,190,190,188],0.719,[188,192,192,203,190,188,190,190,188],21.08,[188,195,192,205,190,188,190,190,188],0.379,[188,198,192,207,190,188,188,190,188],13,[188,188,195,209,190,188,190,190,188],0.795,[188,192,195,211,190,188,190,190,188],23.06,[188,195,195,213,190,188,190,190,188],0.51,[188,198,195,215,190,188,188,190,188],11.2,[192,188,188,217,190,188,190,190,188],0.671,[192,192,188,219,190,188,190,190,188],25.3,[192,195,188,221,190,188,190,190,188],0.371,[192,198,188,223,190,188,188,190,188],11.7,[192,225,188,207,190,188,190,190,188],4,[192,188,192,227,190,188,190,190,188],0.723,[192,192,192,229,190,188,190,190,188],21.72,[192,195,192,231,190,188,190,190,188],0.33,[192,198,192,233,190,188,188,190,188],17.1,[192,225,192,207,190,188,190,190,188],[192,188,195,236,190,188,190,190,188],0.797,[192,192,195,238,190,188,190,190,188],23.62,[192,195,195,240,190,188,190,190,188],0.423,[192,198,195,242,190,188,188,190,188],3.26,[192,225,195,207,190,188,190,190,188],[195,188,188,245,190,188,190,190,188],0.699,[195,192,188,247,190,188,190,190,188],25.59,[195,195,188,249,190,188,190,190,188],0.331,[195,198,188,251,190,188,188,190,188],9.43,[195,225,188,253,190,188,190,190,188],48,[195,188,192,255,190,188,190,190,188],0.745,[195,192,192,257,190,188,190,190,188],21.92,[195,195,192,259,190,188,190,190,188],0.305,[195,198,192,261,190,188,188,190,188],14.4,[195,225,192,253,190,188,190,190,188],[195,188,195,264,190,188,190,190,188],0.817,[195,192,195,266,190,188,190,190,188],24.96,[195,195,195,268,190,188,190,190,188],0.39,[195,198,195,270,190,188,188,190,188],2.79,[195,225,195,253,190,188,190,190,188],[198,198,188,273,190,188,192,190,188],0.06,[198,225,188,275,190,188,190,190,188],8.6,[198,188,192,277,190,188,190,190,188],0.759,[198,192,192,279,190,188,190,190,188],22.22,[198,195,192,281,190,188,190,190,188],0.257,[198,198,192,283,190,188,192,190,188],0.14,[198,225,192,275,190,188,190,190,188],[198,188,195,286,190,188,190,190,188],0.901,[198,192,195,288,190,188,190,190,188],29.4,[198,195,195,290,190,188,190,190,188],0.245,[198,198,195,292,190,188,192,190,188],0.09,[198,225,195,275,190,188,190,190,188],[225,188,188,295,190,188,190,190,188],0.77,[225,192,188,297,190,188,190,190,188],25.6,[225,195,188,299,190,188,190,190,188],0.279,[225,301,188,302,190,188,188,190,188],5,6.42,[225,198,188,304,190,188,188,190,188],160,[225,225,188,306,190,188,190,190,188],523,[225,188,192,308,190,188,190,190,188],0.767,[225,192,192,310,190,188,190,190,188],21.2,[225,195,192,312,190,188,190,190,188],0.28,[225,314,192,315,190,188,188,190,188],6,6.92,[225,198,192,317,190,188,188,190,188],197,[225,225,192,319,190,188,190,190,188],270,[225,188,195,321,190,188,190,190,188],0.875,[225,192,195,323,190,188,190,190,188],27.78,[225,195,195,325,190,188,190,190,188],0.317,[225,327,195,328,190,188,188,190,188],7,4.58,[225,198,195,330,190,188,188,190,188],172,[225,225,195,332,190,188,190,190,188],386,[301,188,188,334,190,188,190,190,188],0.815,[301,192,188,336,190,188,190,190,188],27.21,[301,195,188,338,190,188,190,190,188],0.214,[301,340,188,341,190,188,188,190,188],8,41.55,[301,198,188,343,190,188,188,190,188],134,[301,225,188,345,190,188,190,190,188],734,[301,188,192,347,190,188,190,190,188],0.841,[301,192,192,349,190,188,190,190,188],23.14,[301,195,192,351,190,188,190,190,188],0.183,[301,353,192,354,190,188,188,190,188],9,26.9,[301,198,192,356,190,188,188,190,188],154,[301,225,192,358,190,188,190,190,188],411,[301,188,195,360,190,188,190,190,188],0.903,[301,192,195,362,190,188,190,190,188],29.41,[301,195,195,364,190,188,190,190,188],0.243,[301,366,195,367,190,188,188,190,188],10,36.03,[301,198,195,369,190,188,188,190,188],137,[301,225,195,371,190,188,190,190,188],676,[],[131],[74,375],"not stated for this row",[],[378],"Novel-view synthesis on held-out views (every 8th photo), average per dataset, with training time, rendering FPS and model memory; all on an A6000 GPU except Mip-NeRF360 (trained on a 4-GPU A100 node for 12 h, stated as 48 h single-GPU equivalent); dagger values copied from the Mip-NeRF360 paper",{"slug":380,"group":381,"sourceId":382,"sourceLabel":383,"table":384,"selfRows":385,"metrics":386,"seqs":406,"entrants":427,"cells":436,"outcomes":545,"locators":546,"hardware":547,"wordings":549,"notes":550},"livgaussmap2024-table-iii","livgaussmap2024:Table III","livgaussmap2024","Hong et al., 2024","Table III",18,[387,390,392,394,396,398,400,402,403,404],{"label":388,"unit":139,"statistic":389,"alignment":38},"PSNR [dB] (Interpolated)","not_reported",{"label":391,"unit":139,"statistic":389,"alignment":38},"PSNR [dB] (Extrapolate)",{"label":393,"unit":150,"statistic":389,"alignment":38},"Cost time (written '26m9s')",{"label":395,"unit":150,"statistic":389,"alignment":38},"Cost time (written '16m58s')",{"label":397,"unit":150,"statistic":389,"alignment":38},"Cost time (written '24m16s')",{"label":399,"unit":150,"statistic":389,"alignment":38},"Cost time (written '14m15s')",{"label":401,"unit":150,"statistic":389,"alignment":38},"Cost time (written '20m43s')",{"label":388,"unit":139,"statistic":136,"alignment":38},{"label":391,"unit":139,"statistic":136,"alignment":38},{"label":405,"unit":150,"statistic":136,"alignment":38},"Cost time (written '20m28s')",[407,411,414,417,419,423],{"dataset":408,"sequence":409,"environment":410},"FAST-LIVO dataset","HKU_MB(outdoor)","outdoor, Livox Avia",{"dataset":408,"sequence":412,"environment":413},"HKU_LSK(indoor)","indoor, Livox Avia",{"dataset":415,"sequence":416,"environment":410},"self-collected (Our Device II)","UST_C2_outdoor",{"dataset":415,"sequence":418,"environment":413},"UST_C2_indoor",{"dataset":420,"sequence":421,"environment":422},"self-collected (Our Device I)","UST_RBMS","indoor, RealSense L515",{"dataset":424,"sequence":425,"environment":426},"all five sequences","Avg.","mixed",[428,430,432,434],{"name":429,"methodId":5,"linkable":183,"proposed":67,"self":183},"Case I (3D-GS baseline)",{"name":431,"methodId":77,"linkable":67,"proposed":67,"self":67},"Case II (LiDAR initialization only)",{"name":433,"methodId":77,"linkable":67,"proposed":67,"self":67},"Case III (+ photometric position optimization)",{"name":435,"methodId":382,"linkable":183,"proposed":183,"self":67},"Case IV (full method)",[437,439,441,443,445,447,449,451,453,455,457,459,461,463,465,467,469,471,473,475,477,479,481,483,485,487,489,491,493,495,497,499,501,503,505,507,509,511,513,515,517,519,521,523,525,527,529,531,533,535,537,539,541,543],[188,188,188,438,190,188,190,190,188],24.39,[188,192,188,440,190,188,190,190,188],15.144,[188,195,188,442,190,188,188,190,188],26.15,[188,188,192,444,190,188,190,190,188],31.222,[188,192,192,446,190,188,190,190,188],23.831,[188,198,192,448,190,188,188,190,188],16.97,[188,188,195,450,190,188,190,190,188],31.843,[188,192,195,452,190,188,190,190,188],24.426,[188,225,195,454,190,188,188,190,188],24.27,[188,188,198,456,190,188,190,190,188],31.721,[188,192,198,458,190,188,190,190,188],18.657,[188,301,198,460,190,188,188,190,188],14.25,[188,188,225,462,190,188,190,190,188],31.663,[188,192,225,464,190,188,190,190,188],23.868,[188,314,225,466,190,188,188,190,188],20.72,[188,327,301,468,190,188,190,190,188],30.168,[188,340,301,470,190,188,190,190,188],21.185,[188,353,301,472,190,188,188,190,188],20.47,[192,188,188,474,190,188,190,190,188],24.341,[192,192,188,476,190,188,190,190,188],16.503,[192,188,192,478,190,188,190,190,188],25.964,[192,192,192,480,190,188,190,190,188],22.4,[192,188,195,482,190,188,190,190,188],31.983,[192,192,195,484,190,188,190,190,188],25.653,[192,188,198,486,190,188,190,190,188],29.625,[192,192,198,488,190,188,190,190,188],20.511,[192,188,225,490,190,188,190,190,188],30.211,[192,192,225,492,190,188,190,190,188],23.792,[192,327,301,494,190,188,190,190,188],28.425,[192,340,301,496,190,188,190,190,188],21.772,[195,188,188,498,190,188,190,190,188],24.24,[195,192,188,500,190,188,190,190,188],16.178,[195,188,192,502,190,188,190,190,188],31.045,[195,192,192,504,190,188,190,190,188],24.821,[195,188,195,506,190,188,190,190,188],33.229,[195,192,195,508,190,188,190,190,188],25.047,[195,188,198,510,190,188,190,190,188],31.975,[195,192,198,512,190,188,190,190,188],18.964,[195,188,225,514,190,188,190,190,188],31.047,[195,192,225,516,190,188,190,190,188],24.879,[195,327,301,518,190,188,190,190,188],30.307,[195,340,301,520,190,188,190,190,188],21.978,[198,188,188,522,190,188,190,190,188],25.14,[198,192,188,524,190,188,190,190,188],16.53,[198,188,192,526,190,188,190,190,188],31.597,[198,192,192,528,190,188,190,190,188],24.808,[198,188,195,530,190,188,190,190,188],33.644,[198,192,195,532,190,188,190,190,188],25.912,[198,188,198,534,190,188,190,190,188],32.726,[198,192,198,536,190,188,190,190,188],19.22,[198,188,225,538,190,188,190,190,188],31.277,[198,192,225,540,190,188,190,190,188],25.545,[198,327,301,542,190,188,190,190,188],30.877,[198,340,301,544,190,188,190,190,188],22.403,[],[384],[548],"Intel Core i9 12900K 3.50 GHz, single NVIDIA GeForce RTX 4090",[],[551],"Ablation of map structure optimization: Case I = 3D-GS baseline; Case II = LiDAR-initialized Gaussians without visual structure optimization; Case III = Case II plus photometric position optimization; Case IV = full method with Gaussian pose refinement; SSIM and LPIPS rows of this table omitted here; Cases II and III are ablation variants (method_id null)",{"slug":553,"group":554,"sourceId":555,"sourceLabel":556,"table":131,"selfRows":557,"metrics":558,"seqs":564,"entrants":601,"cells":617,"outcomes":823,"locators":826,"hardware":827,"wordings":829,"notes":830},"huang2024-2dgs-table-1","huang2024_2dgs:Table 1","huang2024_2dgs","Huang et al., 2024a",17,[559,561,562],{"label":560,"unit":389,"statistic":389,"alignment":389},"Chamfer distance (CD)",{"label":560,"unit":389,"statistic":136,"alignment":389},{"label":563,"unit":150,"statistic":389,"alignment":38},"Time (training\u002Freconstruction time)",[565,569,571,573,575,577,579,581,583,585,587,589,591,593,595,597,599],{"dataset":566,"sequence":567,"environment":568},"DTU","scan 24","DTU: 15 scenes with 49 or 69 images at 1600 x 1200, downsampled to 800 x 600 (scene type not described in the paper)",{"dataset":566,"sequence":570,"environment":568},"scan 37",{"dataset":566,"sequence":572,"environment":568},"scan 40",{"dataset":566,"sequence":574,"environment":568},"scan 55",{"dataset":566,"sequence":576,"environment":568},"scan 63",{"dataset":566,"sequence":578,"environment":568},"scan 65",{"dataset":566,"sequence":580,"environment":568},"scan 69",{"dataset":566,"sequence":582,"environment":568},"scan 83",{"dataset":566,"sequence":584,"environment":568},"scan 97",{"dataset":566,"sequence":586,"environment":568},"scan 105",{"dataset":566,"sequence":588,"environment":568},"scan 106",{"dataset":566,"sequence":590,"environment":568},"scan 110",{"dataset":566,"sequence":592,"environment":568},"scan 114",{"dataset":566,"sequence":594,"environment":568},"scan 118",{"dataset":566,"sequence":596,"environment":568},"scan 122",{"dataset":566,"sequence":598,"environment":568},"Mean (15 scans)",{"dataset":566,"sequence":600,"environment":568},"all 15 scans",[602,605,607,609,611,613,615],{"name":603,"methodId":604,"linkable":183,"proposed":67,"self":67},"NeRF (Mildenhall et al., 2021)","nerf2020",{"name":606,"methodId":77,"linkable":67,"proposed":67,"self":67},"VolSDF (Yariv et al., 2021)",{"name":608,"methodId":77,"linkable":67,"proposed":67,"self":67},"NeuS (Wang et al., 2021)",{"name":610,"methodId":5,"linkable":183,"proposed":67,"self":183},"3DGS (Kerbl et al., 2023)",{"name":612,"methodId":77,"linkable":67,"proposed":67,"self":67},"SuGaR (Guédon and Lepetit, 2023)",{"name":614,"methodId":555,"linkable":183,"proposed":183,"self":67},"2DGS-15k (Ours)",{"name":616,"methodId":555,"linkable":183,"proposed":183,"self":67},"2DGS-30k (Ours)",[618,620,622,624,626,628,630,632,634,636,638,640,643,645,647,650,653,655,657,659,661,663,665,667,669,671,673,674,676,678,680,682,684,686,687,688,690,692,694,696,698,700,702,704,706,708,710,712,713,715,717,718,720,722,724,726,728,730,732,734,736,738,740,742,743,745,747,749,750,751,753,755,756,758,760,761,763,765,766,768,770,772,773,774,775,776,778,780,781,783,785,786,787,789,790,792,793,794,795,796,798,799,801,802,804,805,806,808,809,810,811,812,813,814,816,817,818,819,821],[188,188,188,619,190,188,190,190,188],1.9,[188,188,192,621,190,188,190,190,188],1.6,[188,188,195,623,190,188,190,190,188],1.85,[188,188,198,625,190,188,190,190,188],0.58,[188,188,225,627,190,188,190,190,188],2.28,[188,188,301,629,190,188,190,190,188],1.27,[188,188,314,631,190,188,190,190,188],1.47,[188,188,327,633,190,188,190,190,188],1.67,[188,188,340,635,190,188,190,190,188],2.05,[188,188,353,637,190,188,190,190,188],1.07,[188,188,366,639,190,188,190,190,188],0.88,[188,188,641,642,190,188,190,190,188],11,2.53,[188,188,126,644,190,188,190,190,188],1.06,[188,188,207,646,190,188,190,190,188],1.15,[188,188,648,649,190,188,190,190,188],14,0.96,[188,192,651,652,190,188,190,190,188],15,1.49,[188,195,654,77,188,188,188,190,188],16,[192,188,188,656,190,188,190,190,188],1.14,[192,188,192,658,190,188,190,190,188],1.26,[192,188,195,660,190,188,190,190,188],0.81,[192,188,198,662,190,188,190,190,188],0.49,[192,188,225,664,190,188,190,190,188],1.25,[192,188,301,666,190,188,190,190,188],0.7,[192,188,314,668,190,188,190,190,188],0.72,[192,188,327,670,190,188,190,190,188],1.29,[192,188,340,672,190,188,190,190,188],1.18,[192,188,353,666,190,188,190,190,188],[192,188,366,675,190,188,190,190,188],0.66,[192,188,641,677,190,188,190,190,188],1.08,[192,188,126,679,190,188,190,190,188],0.42,[192,188,207,681,190,188,190,190,188],0.61,[192,188,648,683,190,188,190,190,188],0.55,[192,192,651,685,190,188,190,190,188],0.86,[192,195,654,77,188,188,188,190,188],[195,188,188,192,190,188,190,190,188],[195,188,192,689,190,188,190,190,188],1.37,[195,188,195,691,190,188,190,190,188],0.93,[195,188,198,693,190,188,190,190,188],0.43,[195,188,225,695,190,188,190,190,188],1.1,[195,188,301,697,190,188,190,190,188],0.65,[195,188,314,699,190,188,190,190,188],0.57,[195,188,327,701,190,188,190,190,188],1.48,[195,188,340,703,190,188,190,190,188],1.09,[195,188,353,705,190,188,190,190,188],0.83,[195,188,366,707,190,188,190,190,188],0.52,[195,188,641,709,190,188,190,190,188],1.2,[195,188,126,711,190,188,190,190,188],0.35,[195,188,207,662,190,188,190,190,188],[195,188,648,714,190,188,190,190,188],0.54,[195,192,651,716,190,188,190,190,188],0.84,[195,195,654,77,188,188,188,190,188],[198,188,188,719,190,188,190,190,188],2.14,[198,188,192,721,190,188,190,190,188],1.53,[198,188,195,723,190,188,190,190,188],2.08,[198,188,198,725,190,188,190,190,188],1.68,[198,188,225,727,190,188,190,190,188],3.49,[198,188,301,729,190,188,190,190,188],2.21,[198,188,314,731,190,188,190,190,188],1.43,[198,188,327,733,190,188,190,190,188],2.07,[198,188,340,735,190,188,190,190,188],2.22,[198,188,353,737,190,188,190,190,188],1.75,[198,188,366,739,190,188,190,190,188],1.79,[198,188,641,741,190,188,190,190,188],2.55,[198,188,126,721,190,188,190,190,188],[198,188,207,744,190,188,190,190,188],1.52,[198,188,648,746,190,188,190,190,188],1.5,[198,192,651,748,190,188,190,190,188],1.96,[198,195,654,215,190,188,188,190,188],[225,188,188,631,190,188,190,190,188],[225,188,192,752,190,188,190,190,188],1.33,[225,188,195,754,190,188,190,190,188],1.13,[225,188,198,681,190,188,190,190,188],[225,188,225,757,190,188,190,190,188],2.25,[225,188,301,759,190,188,190,190,188],1.71,[225,188,314,646,190,188,190,190,188],[225,188,327,762,190,188,190,190,188],1.63,[225,188,340,764,190,188,190,190,188],1.62,[225,188,353,637,190,188,190,190,188],[225,188,366,767,190,188,190,190,188],0.79,[225,188,641,769,190,188,190,190,188],2.45,[225,188,126,771,190,188,190,190,188],0.98,[225,188,207,639,190,188,190,190,188],[225,188,648,767,190,188,190,190,188],[225,192,651,752,190,188,190,190,188],[225,195,654,77,192,188,188,190,188],[301,188,188,777,190,188,190,190,188],0.48,[301,188,192,779,190,188,190,190,188],0.92,[301,188,195,679,190,188,190,190,188],[301,188,198,782,190,188,190,190,188],0.4,[301,188,225,784,190,188,190,190,188],1.04,[301,188,301,705,190,188,190,190,188],[301,188,314,705,190,188,190,190,188],[301,188,327,788,190,188,190,190,188],1.36,[301,188,340,629,190,188,190,190,188],[301,188,353,791,190,188,190,190,188],0.76,[301,188,366,668,190,188,190,190,188],[301,188,641,762,190,188,190,190,188],[301,188,126,782,190,188,190,190,188],[301,188,207,791,190,188,190,190,188],[301,188,648,797,190,188,190,190,188],0.6,[301,192,651,705,190,188,190,190,188],[301,195,654,800,190,188,188,190,188],5.5,[314,188,188,777,190,188,190,190,188],[314,188,192,803,190,188,190,190,188],0.91,[314,188,195,268,190,188,190,190,188],[314,188,198,268,190,188,190,190,188],[314,188,225,807,190,188,190,190,188],1.01,[314,188,301,705,190,188,190,190,188],[314,188,314,660,190,188,190,190,188],[314,188,327,788,190,188,190,190,188],[314,188,340,629,190,188,190,190,188],[314,188,353,791,190,188,190,190,188],[314,188,366,666,190,188,190,190,188],[314,188,641,815,190,188,190,190,188],1.4,[314,188,126,782,190,188,190,190,188],[314,188,207,791,190,188,190,190,188],[314,188,648,707,190,188,190,190,188],[314,192,651,820,190,188,190,190,188],0.8,[314,195,654,822,190,188,188,190,188],10.9,[824,825],"reported as '>12h' (not a point value)","reported as '~1h' (not a point value)",[131],[828],"2DGS experiments on a single RTX 3090 (written 'GTX RTX3090'); hardware for baselines not stated",[],[831],"Chamfer distance per DTU scan (15 scans) and mean; meshes of 3DGS and 2DGS by TSDF fusion of rendered depth; unit not stated in the paper; images downsampled to 800 x 600; COLMAP sparse points for initialization",{"slug":833,"group":834,"sourceId":382,"sourceLabel":383,"table":835,"selfRows":654,"metrics":836,"seqs":855,"entrants":859,"cells":876,"outcomes":985,"locators":986,"hardware":987,"wordings":988,"notes":989},"livgaussmap2024-table-ii","livgaussmap2024:Table II","Table II",[837,839,841,843,845,847,849,851,853],{"label":838,"unit":139,"statistic":389,"alignment":38},"PSNR (Interpolate)",{"label":840,"unit":38,"statistic":389,"alignment":38},"SSIM (Interpolate)",{"label":842,"unit":38,"statistic":389,"alignment":38},"LPIPS (Interpolate)",{"label":844,"unit":139,"statistic":389,"alignment":38},"PSNR (Extrapolate)",{"label":846,"unit":38,"statistic":389,"alignment":38},"SSIM (Extrapolate)",{"label":848,"unit":38,"statistic":389,"alignment":38},"LPIPS (Extrapolate)",{"label":850,"unit":144,"statistic":389,"alignment":38},"FPS",{"label":852,"unit":150,"statistic":389,"alignment":38},"Cost Time (written '14m10s')",{"label":854,"unit":150,"statistic":389,"alignment":38},"Cost Time (written '8m11s')",[856],{"dataset":857,"sequence":858,"environment":858},"not stated (real-world dataset)","not stated",[860,862,864,866,868,870,872,874],{"name":861,"methodId":77,"linkable":67,"proposed":67,"self":67},"Point-NeRF* [2]",{"name":863,"methodId":77,"linkable":67,"proposed":67,"self":67},"DS-NeRF* [24]",{"name":865,"methodId":5,"linkable":183,"proposed":67,"self":183},"3D-GS* [1]",{"name":867,"methodId":77,"linkable":67,"proposed":67,"self":67},"Plenoxel [18]",{"name":869,"methodId":77,"linkable":67,"proposed":67,"self":67},"M-NeRF360 [16]",{"name":871,"methodId":77,"linkable":67,"proposed":67,"self":67},"F2-NeRF [25]",{"name":873,"methodId":5,"linkable":183,"proposed":67,"self":183},"3D-GS [1]",{"name":875,"methodId":382,"linkable":183,"proposed":183,"self":67},"Our method",[877,879,881,883,885,887,888,889,891,893,895,897,899,901,902,904,906,908,910,912,914,916,918,920,922,924,926,928,930,932,934,936,938,940,942,944,946,948,950,952,954,956,958,960,962,963,965,967,968,970,972,973,975,977,979,980,982,983],[188,188,188,878,190,188,190,190,188],27.331,[188,192,188,880,190,188,190,190,188],0.872,[188,195,188,882,190,188,190,190,188],0.457,[188,198,188,884,190,188,190,190,188],13.117,[188,225,188,886,190,188,190,190,188],0.631,[188,301,188,681,190,188,190,190,188],[188,314,188,273,190,188,188,190,188],[192,188,188,890,190,188,190,190,188],27.178,[192,192,188,892,190,188,190,190,188],0.831,[192,195,188,894,190,188,190,190,188],0.428,[192,198,188,896,190,188,190,190,188],18.534,[192,225,188,898,190,188,190,190,188],0.712,[192,301,188,900,190,188,190,190,188],0.533,[192,314,188,273,190,188,188,190,188],[195,188,188,903,190,188,190,190,188],31.9,[195,192,188,905,190,188,190,190,188],0.913,[195,195,188,907,190,188,190,190,188],0.241,[195,198,188,909,190,188,190,190,188],15.112,[195,225,188,911,190,188,190,190,188],0.647,[195,301,188,913,190,188,190,190,188],0.503,[195,327,188,915,190,188,188,190,188],14.17,[195,314,188,917,190,188,188,190,188],47,[198,188,188,919,190,188,190,190,188],26.744,[198,192,188,921,190,188,190,190,188],0.844,[198,195,188,923,190,188,190,190,188],0.452,[198,198,188,925,190,188,190,190,188],12.916,[198,225,188,927,190,188,190,190,188],0.628,[198,301,188,929,190,188,190,190,188],0.575,[198,314,188,931,190,188,188,190,188],6.12,[225,188,188,933,190,188,190,190,188],28.446,[225,192,188,935,190,188,190,190,188],0.82,[225,195,188,937,190,188,190,190,188],0.444,[225,198,188,939,190,188,190,190,188],19.213,[225,225,188,941,190,188,190,190,188],0.726,[225,301,188,943,190,188,190,190,188],0.526,[225,314,188,945,190,188,188,190,188],0.05,[301,188,188,947,190,188,190,190,188],32.556,[301,192,188,949,190,188,190,190,188],0.941,[301,195,188,951,190,188,190,190,188],0.193,[301,198,188,953,190,188,190,190,188],19.1,[301,225,188,955,190,188,190,190,188],0.764,[301,301,188,957,190,188,190,190,188],0.387,[301,314,188,959,190,188,188,190,188],13.9,[314,188,188,961,190,188,190,190,188],31.899,[314,192,188,905,190,188,190,190,188],[314,195,188,964,190,188,190,190,188],0.24,[314,198,188,966,190,188,190,190,188],15.111,[314,225,188,911,190,188,190,190,188],[314,301,188,969,190,188,190,190,188],0.502,[314,340,188,971,190,188,188,190,188],8.18,[314,314,188,114,190,188,188,190,188],[327,188,188,974,190,188,190,190,188],32.787,[327,192,188,976,190,188,190,190,188],0.926,[327,195,188,978,190,188,190,190,188],0.19,[327,198,188,536,190,188,190,190,188],[327,225,188,981,190,188,190,190,188],0.803,[327,301,188,249,190,188,190,190,188],[327,314,188,984,190,188,188,190,188],43,[],[835],[548],[],[990],"Novel-view synthesis on interpolated and extrapolated views on a real-world dataset (dataset and sequence not named for this table); asterisk methods were enhanced with dense LiDAR point clouds; cost time and FPS on the authors' desktop",[992,1001,1007,1012,1021,1028,1034,1040],{"group":993,"slug":994,"sourceLabel":995,"table":131,"selfRows":126,"datasets":996},"gslivm2025:Table 1","gslivm2025-table-1","Xie et al., 2025",[997,998,999,1000],"Botanic Garden","FAST-LIVO dataset (row-to-dataset mapping inferred from caption order)","NTU-VIRAL","R3LIVE dataset",{"group":1002,"slug":1003,"sourceLabel":556,"table":1004,"selfRows":340,"datasets":1005},"huang2024_2dgs:Table 2","huang2024-2dgs-table-2","Table 2",[1006],"Tanks and Temples",{"group":1008,"slug":1009,"sourceLabel":995,"table":1010,"selfRows":314,"datasets":1011},"gslivm2025:Supp. Table 4","gslivm2025-supp-table-4","Supp. Table 4",[997,1000],{"group":1013,"slug":1014,"sourceLabel":1015,"table":1016,"selfRows":314,"datasets":1017},"gslivo2025:Table I","gslivo2025-table-i","Hong et al., 2025","Table I",[1018,1019,1020],"FAST-LIVO2 dataset","MARS-LVIG","proprietary (MoCap)",{"group":1022,"slug":1023,"sourceLabel":995,"table":1024,"selfRows":301,"datasets":1025},"gslivm2025:Supp. Table 5","gslivm2025-supp-table-5","Supp. Table 5",[997,1026,858,1027],"FAST-LIVO or R3LIVE dataset (not stated)","self-collected",{"group":1029,"slug":1030,"sourceLabel":383,"table":1031,"selfRows":198,"datasets":1032},"livgaussmap2024:Table IV","livgaussmap2024-table-iv","Table IV",[1033],"FusionPortable",{"group":1035,"slug":1036,"sourceLabel":995,"table":1037,"selfRows":195,"datasets":1038},"gslivm2025:ICCV Supp. Table 5","gslivm2025-iccv-supp-table-5","ICCV Supp. Table 5",[1039,1000],"FAST-LIVO dataset (row-to-dataset mapping inferred from Table 1 caption order)",{"group":1041,"slug":1042,"sourceLabel":995,"table":1043,"selfRows":192,"datasets":1044},"gslivm2025:ICCV Supp. Table 6","gslivm2025-iccv-supp-table-6","ICCV Supp. Table 6",[858],1790510664595]