[{"data":1,"prerenderedAt":769},["ShallowReactive",2],{"method-mast3r2024":3},{"method":4,"reference":51,"equipment":71,"figures":93,"results":94},{"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":26,"sensors":33,"platform":35,"estimator":36,"association":37,"timeModel":38,"deskew":38,"loopClosure":38,"globalOptimization":39,"mapRepresentation":40,"prior":41,"outputGeometry":42,"compute":43,"codeUrl":44,"codeLicense":45,"relatedVersions":46},"mast3r2024","Leroy et al., 2024","MASt3R","Grounding Image Matching in 3D with MASt3R",2024,"recent","C09","registration_component","MASt3R 在 DUSt3R 上增加輸出稠密局部特徵的分支並以匹配損失訓練，同時提出快速互為最近鄰匹配以降低二次複雜度。與 DUSt3R 不同，當訓練真值為公制時不做尺度正規化，使模型可輸出公制尺度點圖。其 DTU 結果是以真值相機對匹配點三角化取得，並非完全無相機的重建。","Adds a dense feature head and metric-aware regression to DUSt3R for 3D-grounded matching; the DTU MVS numbers use GT cameras for triangulation.","full_text_reviewed","peer_reviewed_published","background","論文未涉及營建場域。",[20],"public_benchmark",[22,23,24,25],"Map-free test: VCRE AUC 0.933 and median translation error 0.36 m using MASt3R's own metric depth, vs 0.697 and 0.97 m for DUSt3R (Table 2, VoR)","Zero-shot DTU MVS overall Chamfer 0.374 mm vs 1.741 mm for DUSt3R and within the range of DTU-trained methods (0.295 to 0.462 mm) (Table 3 right, VoR)","InLoc: top-40 retrieval reaches 56.1\u002F79.3\u002F90.9% (DUC1) and 71.0\u002F87.0\u002F91.6% (DUC2), above the listed baselines (Table 4)","Fast reciprocal matching speeds matching and improves pose accuracy through more uniform match coverage (Sec. 3.3; arXiv v1 App. B)",[27,28,29,30,31,32],"DTU reconstruction relies on GT cameras for triangulation (DTU paragraph)","Follow-up [mast3rslam2025] found scale often inconsistent across MASt3R predictions (MASt3R-SLAM Sec. 3.1)","Network input limited to 512 px on the largest side; coarse-only matching nearly doubles DTU errors (Sec. 3.4; arXiv v1 App. C Table 5)","Direct pointmap regression (PnP on the pointmap) gives poor localization in larger scenes such as Aachen and InLoc (Sec. 4.4, Table 4)","Pairwise only in this paper; DUSt3R's multi-view global alignment is not used (Sec. 3.1)","(reviewer inference) The VoR text still reports a 30-point VCRE AUC gain over LoFTR+KBR (0.634), but the VoR Table 2 also lists Mickey at 0.748, so the margin over the best listed baseline is about 19 points",[34],"monocular camera (image pairs)",[],"feed-forward pointmap regression (DUSt3R backbone) with an added dense local-feature head","dense learned local features with fast reciprocal nearest-neighbour matching","not_applicable","none","pairwise pointmaps with confidence and dense descriptors","Learned prior from a mixture of 14 training datasets (10 with metric ground truth), initialized from the public DUSt3R checkpoint (ViT-Large encoder, ViT-Base decoder); regression normalization dropped when ground truth is metric (Sec. 3.1, 4.1)","pointmaps (metric-scale when trained on metric data), dense matches; DTU point clouds by triangulating matches with GT cameras","No GPU or runtime hardware reported; matching time is reported only on a single CPU core (Fig. 2 right in the VoR); fast reciprocal matching with k = 3000 speeds matching about 64 times on Map-free; the network handles at most 512 px on the largest side, so high-resolution images need coarse-to-fine window matching (Sec. 3.3, 3.4, 4.2)","https:\u002F\u002Fgithub.com\u002Fnaver\u002Fmast3r","CC BY-NC-SA 4.0",[47],{"relation":48,"title":49,"doi_or_url":50},"preprint","arXiv:2406.09756","https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.09756",{"id":5,"kind":52,"shortName":7,"title":8,"authors":53,"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":38,"codeUrl":44,"cluster":11,"topics":66,"mdpi":67,"verification":68,"label":6,"fulltextRoute":69,"versionRead":70,"addedByCensus":67},"component",[54,55,56],"Vincent Leroy","Yohann Cabon","Jerome Revaud","Computer Vision - ECCV 2024 (Lecture Notes in Computer Science)","conference","Springer","LNCS, pp. 71-91 (Crossref published-print 2025)","10.1007\u002F978-3-031-73220-1_5","2406.09756","https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1007\u002F978-3-031-73220-1_5","2024-06-14","metadata_verified",[11],false,"corrected","NTU institutional (curl)","Version of record (ECCV 2024, LNCS 15130, pp. 71-91, Springer PDF, 21 pp.) plus the arXiv v1 (2024-06-14) appendix, which corresponds to the VoR supplementary material",[72,79,83,86],{"category":73,"model":74,"canonical":74,"role":75,"dataset":76,"specs":77,"locator":78},"camera","iPhone 7","dataset sensor","InLoc","329 InLoc query images","Sec. 4.4",{"category":73,"model":80,"canonical":80,"role":75,"dataset":81,"specs":82,"locator":78},"mobile phones (models not named)","Aachen Day-Night","824 daytime and 98 nighttime Aachen query images",{"category":73,"model":84,"canonical":84,"role":75,"dataset":81,"specs":85,"locator":78},"hand-held cameras (models not named)","4,328 Aachen reference images",{"category":87,"model":88,"canonical":88,"role":89,"dataset":90,"specs":91,"locator":92},"compute","single CPU core (model not reported)","compute for runtime","Map-free relocalization","platform for the matching-time versus accuracy plot","Fig. 2 right (VoR); Fig. 3 right (arXiv v1)",[],{"totalRows":95,"groupCount":96,"groups":97,"others":743},59,9,[98,326,539,605],{"slug":99,"group":100,"sourceId":5,"sourceLabel":6,"table":101,"selfRows":102,"metrics":103,"seqs":125,"entrants":129,"cells":155,"outcomes":319,"locators":320,"hardware":322,"wordings":323,"notes":324},"mast3r2024-table-2","mast3r2024:Table 2","Table 2",21,[104,108,111,114,118,121,123],{"label":105,"unit":106,"statistic":107,"alignment":39},"VCRE Reproj.","px","not_reported",{"label":109,"unit":110,"statistic":107,"alignment":39},"VCRE Prec. (\u003C 90 px)","%",{"label":112,"unit":113,"statistic":107,"alignment":39},"VCRE AUC (\u003C 90 px)","ratio",{"label":115,"unit":116,"statistic":117,"alignment":39},"Pose Median Err. (translation)","m","median",{"label":119,"unit":120,"statistic":117,"alignment":39},"Pose Median Err. (rotation)","deg",{"label":122,"unit":110,"statistic":107,"alignment":39},"Pose Precision (\u003C 25 cm, 5 deg)",{"label":124,"unit":113,"statistic":107,"alignment":39},"Pose AUC (\u003C 25 cm, 5 deg)",[126],{"dataset":90,"sequence":127,"environment":128},"test set (130 scenes)","Map-free relocalization test set (130 scenes, each with two video sequences); metric relative pose from a single reference image; qualitative pairs with viewpoint changes up to 180 deg",[130,133,135,137,139,141,143,145,149,151,153],{"name":131,"methodId":132,"linkable":67,"proposed":67,"self":67},"RPR [5] (DPT depth)",null,{"name":134,"methodId":132,"linkable":67,"proposed":67,"self":67},"SIFT [54] (DPT depth)",{"name":136,"methodId":132,"linkable":67,"proposed":67,"self":67},"SP+SG [78] (DPT depth)",{"name":138,"methodId":132,"linkable":67,"proposed":67,"self":67},"LoFTR [87] (KBR depth)",{"name":140,"methodId":132,"linkable":67,"proposed":67,"self":67},"FAR [75] (auto)",{"name":142,"methodId":132,"linkable":67,"proposed":67,"self":67},"RoMa [29] (DPT depth)",{"name":144,"methodId":132,"linkable":67,"proposed":67,"self":67},"Mickey [8] (auto)",{"name":146,"methodId":147,"linkable":148,"proposed":67,"self":67},"DUSt3R [106] (DPT depth)","dust3r2024",true,{"name":150,"methodId":5,"linkable":148,"proposed":148,"self":148},"MASt3R (DPT depth)",{"name":152,"methodId":5,"linkable":148,"proposed":148,"self":148},"MASt3R (auto, own metric depth)",{"name":154,"methodId":5,"linkable":148,"proposed":148,"self":148},"MASt3R (direct reg., PnP on pointmap)",[156,160,163,166,169,172,175,177,179,181,183,185,187,189,191,193,195,197,199,201,203,205,207,209,211,213,215,217,219,221,223,225,227,229,231,233,235,237,239,241,243,245,247,249,251,253,255,257,259,261,264,266,268,270,272,274,276,279,281,283,285,287,289,291,293,295,297,299,300,302,304,307,309,311,313,315,317],[157,157,157,158,159,157,159,159,157],0,147.1,-1,[157,161,157,162,159,157,159,159,157],1,40.2,[157,164,157,165,159,157,159,159,157],2,0.402,[157,167,157,168,159,157,159,159,157],3,1.68,[157,170,157,171,159,157,159,159,157],4,22.5,[157,173,157,174,159,157,159,159,157],5,6,[157,174,157,176,159,157,159,159,157],0.06,[161,157,157,178,159,157,159,159,157],222.8,[161,161,157,180,159,157,159,159,157],25,[161,164,157,182,159,157,159,159,157],0.504,[161,167,157,184,159,157,159,159,157],2.93,[161,170,157,186,159,157,159,159,157],61.4,[161,173,157,188,159,157,159,159,157],10.3,[161,174,157,190,159,157,159,159,157],0.252,[164,157,157,192,159,157,159,159,157],160.3,[164,161,157,194,159,157,159,159,157],36.1,[164,164,157,196,159,157,159,159,157],0.602,[164,167,157,198,159,157,159,159,157],1.88,[164,170,157,200,159,157,159,159,157],25.4,[164,173,157,202,159,157,159,159,157],16.8,[164,174,157,204,159,157,159,159,157],0.346,[167,157,157,206,159,157,159,159,157],165,[167,161,157,208,159,157,159,159,157],34.3,[167,164,157,210,159,157,159,159,157],0.634,[167,167,157,212,159,157,159,159,157],2.23,[167,170,157,214,159,157,159,159,157],37.8,[167,173,157,216,159,157,159,159,157],11,[167,174,157,218,159,157,159,159,157],0.295,[170,157,157,220,159,157,159,159,157],137,[170,161,157,222,159,157,159,159,157],44.2,[170,164,157,224,159,157,159,159,157],0.68,[170,167,157,226,159,157,159,159,157],1.48,[170,170,157,228,159,157,159,159,157],17.2,[170,173,157,230,159,157,159,159,157],17.7,[170,174,157,232,159,157,159,159,157],0.392,[173,157,157,234,159,157,159,159,157],128.8,[173,161,157,236,159,157,159,159,157],45.6,[173,164,157,238,159,157,159,159,157],0.669,[173,167,157,240,159,157,159,159,157],1.23,[173,170,157,242,159,157,159,159,157],11.1,[173,173,157,244,159,157,159,159,157],22.8,[173,174,157,246,159,157,159,159,157],0.407,[174,157,157,248,159,157,159,159,157],129.5,[174,161,157,250,159,157,159,159,157],49.3,[174,164,157,252,159,157,159,159,157],0.748,[174,167,157,254,159,157,159,159,157],1.66,[174,170,157,256,159,157,159,159,157],27.3,[174,173,157,258,159,157,159,159,157],13.3,[174,174,157,260,159,157,159,159,157],0.325,[262,157,157,263,159,157,159,159,157],7,116,[262,161,157,265,159,157,159,159,157],50.3,[262,164,157,267,159,157,159,159,157],0.697,[262,167,157,269,159,157,159,159,157],0.97,[262,170,157,271,159,157,159,159,157],7.1,[262,173,157,273,159,157,159,159,157],21.6,[262,174,157,275,159,157,159,159,157],0.394,[277,157,157,278,159,157,159,159,157],8,104,[277,161,157,280,159,157,159,159,157],54.2,[277,164,157,282,159,157,159,159,157],0.726,[277,167,157,284,159,157,159,159,157],0.8,[277,170,157,286,159,157,159,159,157],2.2,[277,173,157,288,159,157,159,159,157],27,[277,174,157,290,159,157,159,159,157],0.456,[96,157,157,292,159,157,159,159,157],48.7,[96,161,157,294,159,157,159,159,157],79.3,[96,164,157,296,159,157,159,159,157],0.933,[96,167,157,298,159,157,159,159,157],0.36,[96,170,157,286,159,157,159,159,157],[96,173,157,301,159,157,159,159,157],54.7,[96,174,157,303,159,157,159,159,157],0.74,[305,157,157,306,159,157,159,159,157],10,53.2,[305,161,157,308,159,157,159,159,157],79.1,[305,164,157,310,159,157,159,159,157],0.941,[305,167,157,312,159,157,159,159,157],0.42,[305,170,157,314,159,157,159,159,157],3.1,[305,173,157,316,159,157,159,159,157],53,[305,174,157,318,159,157,159,159,157],0.777,[],[321],"Table 2 (VoR)",[],[],[325],"Map-free relocalization test set (VoR table, which adds FAR, RoMa and Mickey compared with arXiv v1). VCRE = virtual correspondence reprojection error, precision and AUC at VCRE \u003C 90 px; pose precision and AUC at \u003C 25 cm and 5 deg; median translation and rotation error; the depth column gives the metric-scale source (DPT fine-tuned on KITTI, KBR, or MASt3R's own depth, 'auto').",{"slug":327,"group":328,"sourceId":329,"sourceLabel":330,"table":331,"selfRows":332,"metrics":333,"seqs":343,"entrants":364,"cells":375,"outcomes":528,"locators":533,"hardware":534,"wordings":536,"notes":537},"slam3r2025-table-1","slam3r2025:Table 1","slam3r2025","Liu et al., 2025","Table 1",17,[334,339,341],{"label":335,"unit":336,"statistic":337,"alignment":338},"Acc.","cm","mean","other: Umeyama similarity alignment followed by ICP to the ground-truth point cloud",{"label":340,"unit":336,"statistic":337,"alignment":338},"Comp.",{"label":342,"unit":342,"statistic":337,"alignment":38},"FPS",[344,348,350,352,354,356,358,360,362],{"dataset":345,"sequence":346,"environment":347},"7-Scenes","Chess","real indoor rooms",{"dataset":345,"sequence":349,"environment":347},"Fire",{"dataset":345,"sequence":351,"environment":347},"Heads",{"dataset":345,"sequence":353,"environment":347},"Office",{"dataset":345,"sequence":355,"environment":347},"Pumpkin",{"dataset":345,"sequence":357,"environment":347},"RedKitchen",{"dataset":345,"sequence":359,"environment":347},"Stairs",{"dataset":345,"sequence":361,"environment":347},"average of test sequences",{"dataset":345,"sequence":363,"environment":347},"all test sequences",[365,367,369,371,373],{"name":366,"methodId":147,"linkable":148,"proposed":67,"self":67},"DUSt3R [ 64 ]",{"name":368,"methodId":5,"linkable":148,"proposed":67,"self":148},"MASt3R [ 28 ]",{"name":370,"methodId":132,"linkable":67,"proposed":67,"self":67},"Spann3R [ 61 ]",{"name":372,"methodId":132,"linkable":67,"proposed":148,"self":67},"SLAM3R-NoConf (Ours)",{"name":374,"methodId":132,"linkable":67,"proposed":148,"self":67},"SLAM3R (Ours)",[376,378,380,382,384,385,387,389,391,393,395,397,399,401,403,405,407,408,410,412,414,416,418,419,421,423,425,427,429,430,431,433,435,437,438,439,440,442,444,446,447,449,451,453,455,457,459,461,463,465,467,468,469,471,473,474,475,477,479,481,483,485,487,488,490,492,494,496,497,499,501,503,505,507,509,511,512,514,516,518,520,522,524,525,527],[157,157,157,377,159,157,159,159,157],2.26,[157,161,157,379,159,157,159,159,157],2.13,[157,157,161,381,159,157,159,159,157],1.04,[157,161,161,383,159,157,159,159,157],1.5,[157,157,164,254,159,157,159,159,157],[157,161,164,386,159,157,159,159,157],0.98,[157,157,167,388,159,157,159,159,157],4.62,[157,161,167,390,159,157,159,159,157],4.74,[157,157,170,392,159,157,159,159,157],1.73,[157,161,170,394,159,157,159,159,157],2.43,[157,157,173,396,159,157,159,159,157],1.95,[157,161,173,398,159,157,159,159,157],2.36,[157,157,174,400,159,157,159,159,157],3.37,[157,161,174,402,159,157,159,159,157],10.75,[157,157,262,404,159,157,159,159,157],2.19,[157,161,262,406,159,157,159,159,157],3.24,[157,164,277,132,157,157,157,159,157],[161,157,157,409,159,157,159,159,157],2.08,[161,161,157,411,159,157,159,159,157],2.12,[161,157,161,413,159,157,159,159,157],1.54,[161,161,161,415,159,157,159,159,157],1.43,[161,157,164,417,159,157,159,159,157],1.06,[161,161,164,381,159,157,159,159,157],[161,157,167,420,159,157,159,159,157],3.23,[161,161,167,422,159,157,159,159,157],3.19,[161,157,170,424,159,157,159,159,157],5.68,[161,161,170,426,159,157,159,159,157],3.07,[161,157,173,428,159,157,159,159,157],3.5,[161,161,173,400,159,157,159,159,157],[161,157,174,398,159,157,159,159,157],[161,161,174,432,159,157,159,159,157],13.16,[161,157,262,434,159,157,159,159,157],3.04,[161,161,262,436,159,157,159,159,157],3.9,[161,164,277,132,161,157,157,159,157],[164,157,157,212,159,157,159,159,157],[164,161,157,168,159,157,159,159,157],[164,157,161,441,159,157,159,159,157],0.88,[164,161,161,443,159,157,159,159,157],0.92,[164,157,164,445,159,157,159,159,157],2.67,[164,161,164,386,159,157,159,159,157],[164,157,167,448,159,157,159,159,157],5.86,[164,161,167,450,159,157,159,159,157],3.54,[164,157,170,452,159,157,159,159,157],2.25,[164,161,170,454,159,157,159,159,157],1.85,[164,157,173,456,159,157,159,159,157],2.68,[164,161,173,458,159,157,159,159,157],1.8,[164,157,174,460,159,157,159,159,157],5.65,[164,161,174,462,159,157,159,159,157],5.15,[164,157,262,464,159,157,159,159,157],3.42,[164,161,262,466,159,157,159,159,157],2.41,[164,164,277,132,164,157,157,159,157],[167,157,157,411,159,157,159,159,157],[167,161,157,470,159,157,159,159,157],1.21,[167,157,161,472,159,157,159,159,157],0.95,[167,161,161,284,159,157,159,159,157],[167,157,164,420,159,157,159,159,157],[167,161,164,476,159,157,159,159,157],1.67,[167,157,167,478,159,157,159,159,157],2.59,[167,161,167,480,159,157,159,159,157],2.21,[167,157,170,482,159,157,159,159,157],1.99,[167,161,170,484,159,157,159,159,157],2.04,[167,157,173,486,159,157,159,159,157],2.09,[167,161,173,198,159,157,159,159,157],[167,157,174,489,159,157,159,159,157],4.54,[167,161,174,491,159,157,159,159,157],6.38,[167,157,262,493,159,157,159,159,157],2.4,[167,161,262,495,159,157,159,159,157],2.24,[167,164,277,180,167,157,157,159,157],[170,157,157,498,159,157,159,159,157],1.63,[170,161,157,500,159,157,159,159,157],1.31,[170,157,161,502,159,157,159,159,157],0.84,[170,161,161,504,159,157,159,159,157],0.83,[170,157,164,506,159,157,159,159,157],2.95,[170,161,164,508,159,157,159,159,157],1.22,[170,157,167,510,159,157,159,159,157],2.32,[170,161,167,377,159,157,159,159,157],[170,157,170,513,159,157,159,159,157],1.81,[170,161,170,515,159,157,159,159,157],2.05,[170,157,173,517,159,157,159,159,157],1.84,[170,161,173,519,159,157,159,159,157],1.94,[170,157,174,521,159,157,159,159,157],4.19,[170,161,174,523,159,157,159,159,157],6.91,[170,157,262,379,159,157,159,159,157],[170,161,262,526,159,157,159,159,157],2.34,[170,164,277,180,167,157,157,159,157],[529,530,531,532],"other: as written '\u003C 1'","other: as written '\u003C\u003C 1'","other: as written '> 50'","other: approximate, as written '~25'",[331],[535],"single NVIDIA 4090D GPU (Sec. 4.1)",[],[538],"7 Scenes, one-twentieth of frames of each test sequence as input video; accuracy and completeness in cm against back-projected ground-truth depth; SLAM3R filters points with confidence threshold 3, SLAM3R-NoConf keeps all",{"slug":540,"group":541,"sourceId":542,"sourceLabel":543,"table":544,"selfRows":170,"metrics":545,"seqs":553,"entrants":558,"cells":566,"outcomes":596,"locators":597,"hardware":598,"wordings":600,"notes":601},"vggt2025-table-3","vggt2025:Table 3","vggt2025","Wang et al., 2025b","Table 3",[546,547,548,550],{"label":335,"unit":107,"statistic":107,"alignment":107},{"label":340,"unit":107,"statistic":107,"alignment":107},{"label":549,"unit":107,"statistic":107,"alignment":107},"Overall (Chamfer)",{"label":551,"unit":552,"statistic":107,"alignment":107},"Time (approximate, per reconstruction)","s",[554],{"dataset":555,"sequence":556,"environment":557},"ETH3D","10 random frames per scene","not described in the paper",[559,561,562,564],{"name":560,"methodId":147,"linkable":148,"proposed":67,"self":67},"DUSt3R",{"name":7,"methodId":5,"linkable":148,"proposed":67,"self":148},{"name":563,"methodId":542,"linkable":148,"proposed":148,"self":67},"Ours (Point)",{"name":565,"methodId":542,"linkable":148,"proposed":148,"self":67},"Ours (Depth + Cam)",[567,569,571,573,574,576,578,580,581,583,585,587,589,591,593,595],[157,157,157,568,159,157,159,159,157],1.167,[157,161,157,570,159,157,159,159,157],0.842,[157,164,157,572,159,157,159,159,157],1.005,[157,167,157,262,159,157,157,159,157],[161,157,157,575,159,157,159,159,157],0.968,[161,161,157,577,159,157,159,159,157],0.684,[161,164,157,579,159,157,159,159,157],0.826,[161,167,157,96,159,157,157,159,157],[164,157,157,582,159,157,159,159,161],0.901,[164,161,157,584,159,157,159,159,161],0.518,[164,164,157,586,159,157,159,159,161],0.709,[164,167,157,588,159,157,157,159,161],0.2,[167,157,157,590,159,157,159,159,164],0.873,[167,161,157,592,159,157,159,159,164],0.482,[167,164,157,594,159,157,159,159,164],0.677,[167,167,157,588,159,157,157,159,164],[],[544],[599],"not stated for Table 3 (Table 1 timings used one NVIDIA H100)",[],[602,603,604],"Point map estimation on ETH3D, 10 random frames per scene, predicted cloud aligned to GT with the Umeyama algorithm (similarity or rigid not stated), invalid points filtered with official masks; global alignment; units not stated","Point map estimation on ETH3D, 10 random frames per scene, predicted cloud aligned to GT with the Umeyama algorithm (similarity or rigid not stated), invalid points filtered with official masks; point map head, feed-forward; units not stated","Point map estimation on ETH3D, 10 random frames per scene, predicted cloud aligned to GT with the Umeyama algorithm (similarity or rigid not stated), invalid points filtered with official masks; depth head unprojected with camera head, feed-forward; units not stated",{"slug":606,"group":607,"sourceId":5,"sourceLabel":6,"table":608,"selfRows":167,"metrics":609,"seqs":614,"entrants":619,"cells":650,"outcomes":736,"locators":737,"hardware":739,"wordings":740,"notes":741},"mast3r2024-table-3-right","mast3r2024:Table 3 right","Table 3 right",[610,612,613],{"label":335,"unit":611,"statistic":337,"alignment":39},"mm",{"label":340,"unit":611,"statistic":337,"alignment":39},{"label":549,"unit":611,"statistic":337,"alignment":39},[615],{"dataset":616,"sequence":617,"environment":618},"DTU","evaluation set (average)","DTU MVS evaluation set, zero-shot (no DTU training); input images 1200 x 1600",[620,622,624,626,628,630,632,634,636,638,640,642,644,646,648],{"name":621,"methodId":132,"linkable":67,"proposed":67,"self":67},"Camp [14] (c)",{"name":623,"methodId":132,"linkable":67,"proposed":67,"self":67},"Furu [32] (c)",{"name":625,"methodId":132,"linkable":67,"proposed":67,"self":67},"Tola [95] (c)",{"name":627,"methodId":132,"linkable":67,"proposed":67,"self":67},"Gipuma [33] (c)",{"name":629,"methodId":132,"linkable":67,"proposed":67,"self":67},"MVSNet [114] (d)",{"name":631,"methodId":132,"linkable":67,"proposed":67,"self":67},"CVP-MVSNet [113] (d)",{"name":633,"methodId":132,"linkable":67,"proposed":67,"self":67},"UCS-Net [18] (d)",{"name":635,"methodId":132,"linkable":67,"proposed":67,"self":67},"CER-MVS [57] (d)",{"name":637,"methodId":132,"linkable":67,"proposed":67,"self":67},"CIDER [111] (d)",{"name":639,"methodId":132,"linkable":67,"proposed":67,"self":67},"PatchmatchNet [103] (d)",{"name":641,"methodId":132,"linkable":67,"proposed":67,"self":67},"CasMVSNet [36] (d)",{"name":643,"methodId":132,"linkable":67,"proposed":67,"self":67},"TransMVSNet [22] (d)",{"name":645,"methodId":132,"linkable":67,"proposed":67,"self":67},"GeoMVSNet [122] (d)",{"name":647,"methodId":147,"linkable":148,"proposed":67,"self":67},"DUSt3R [106] (e)",{"name":649,"methodId":5,"linkable":148,"proposed":148,"self":148},"MASt3R (e)",[651,653,655,657,659,660,661,663,665,667,669,670,672,674,676,678,680,682,684,686,688,690,692,694,696,698,700,702,703,705,707,708,710,712,714,716,717,720,722,723,726,728,730,733,734],[157,157,157,652,159,157,159,159,157],0.835,[157,161,157,654,159,157,159,159,157],0.554,[157,164,157,656,159,157,159,159,157],0.695,[161,157,157,658,159,157,159,159,157],0.613,[161,161,157,310,159,157,159,159,157],[161,164,157,318,159,157,159,159,157],[164,157,157,662,159,157,159,159,157],0.342,[164,161,157,664,159,157,159,159,157],1.19,[164,164,157,666,159,157,159,159,157],0.766,[167,157,157,668,159,157,159,159,157],0.283,[167,161,157,590,159,157,159,159,157],[167,164,157,671,159,157,159,159,157],0.578,[170,157,157,673,159,157,159,159,157],0.396,[170,161,157,675,159,157,159,159,157],0.527,[170,164,157,677,159,157,159,159,157],0.462,[173,157,157,679,159,157,159,159,157],0.296,[173,161,157,681,159,157,159,159,157],0.406,[173,164,157,683,159,157,159,159,157],0.351,[174,157,157,685,159,157,159,159,157],0.338,[174,161,157,687,159,157,159,159,157],0.349,[174,164,157,689,159,157,159,159,157],0.344,[262,157,157,691,159,157,159,159,157],0.359,[262,161,157,693,159,157,159,159,157],0.305,[262,164,157,695,159,157,159,159,157],0.332,[277,157,157,697,159,157,159,159,157],0.417,[277,161,157,699,159,157,159,159,157],0.437,[277,164,157,701,159,157,159,159,157],0.427,[96,157,157,701,159,157,159,159,157],[96,161,157,704,159,157,159,159,157],0.277,[96,164,157,706,159,157,159,159,157],0.352,[305,157,157,260,159,157,159,159,157],[305,161,157,709,159,157,159,159,157],0.385,[305,164,157,711,159,157,159,159,157],0.355,[216,157,157,713,159,157,159,159,157],0.321,[216,161,157,715,159,157,159,159,157],0.289,[216,164,157,693,159,157,159,159,157],[718,157,157,719,159,157,159,159,157],12,0.331,[718,161,157,721,159,157,159,159,157],0.259,[718,164,157,218,159,157,159,159,157],[724,157,157,725,159,157,159,159,157],13,2.677,[724,161,157,727,159,157,159,159,157],0.805,[724,164,157,729,159,157,159,159,157],1.741,[731,157,157,732,159,157,159,159,157],14,0.403,[731,161,157,689,159,157,159,159,157],[731,164,157,735,159,157,159,159,157],0.374,[],[738],"Table 3 right (VoR)",[],[],[742],"DTU dense MVS (mm): accuracy, completeness and overall Chamfer (average of the two) with the benchmark's evaluation code; MASt3R and DUSt3R zero-shot; MASt3R matches are computed without camera knowledge but triangulated with ground-truth cameras in the ground-truth frame, then filtered by geometric consistency. Groups: (c) handcrafted, (d) learning-based trained on DTU, (e) zero-shot. VoR adds CasMVSNet and TransMVSNet compared with arXiv v1.",[744,749,754,761,765],{"group":745,"slug":746,"sourceLabel":6,"table":747,"selfRows":167,"datasets":748},"mast3r2024:Table 5 right (arXiv v1 App. C)","mast3r2024-table-5-right-arxiv-v1-app-c","Table 5 right (arXiv v1 App. C)",[616],{"group":750,"slug":751,"sourceLabel":330,"table":101,"selfRows":167,"datasets":752},"slam3r2025:Table 2","slam3r2025-table-2",[753],"Replica",{"group":755,"slug":756,"sourceLabel":543,"table":331,"selfRows":167,"datasets":757},"vggt2025:Table 1","vggt2025-table-1",[758,759,760],"CO3Dv2","RealEstate10K (unseen)","RealEstate10K and CO3Dv2 (single time column)",{"group":762,"slug":763,"sourceLabel":543,"table":101,"selfRows":167,"datasets":764},"vggt2025:Table 2","vggt2025-table-2",[616],{"group":766,"slug":767,"sourceLabel":330,"table":544,"selfRows":164,"datasets":768},"slam3r2025:Table 3","slam3r2025-table-3",[345,753],1790510665294]