[{"data":1,"prerenderedAt":470},["ShallowReactive",2],{"method-sdvloam2023":3},{"method":4,"reference":55,"equipment":79,"figures":113,"results":114},{"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":28,"sensors":33,"platform":36,"estimator":39,"association":40,"timeModel":41,"deskew":42,"loopClosure":43,"globalOptimization":44,"mapRepresentation":45,"prior":46,"outputGeometry":47,"compute":48,"codeUrl":49,"codeLicense":50,"relatedVersions":51},"sdvloam2023","Yuan et al., 2023a","SDV-LOAM","SDV-LOAM: Semi-Direct Visual-LiDAR Odometry and Mapping",2023,"recent","C07","odometry_with_local_mapping","SDV-LOAM 把視覺與 LiDAR 分成前後兩個模組：視覺模組是半直接法深度增強視覺里程計，先以光度誤差直接估計位姿，再做帶傳播的點匹配與重投影修正，並以滑動視窗光束法平差最佳化，追蹤點的深度直接取自投影的 LiDAR 點；其位姿作為 LiDAR 模組的運動先驗。LiDAR 模組以 CT-ICP 為基礎，將 10 Hz 掃描重組成與 60 Hz 影像同步的片段，並依地面與垂直約束比例在 3 自由度與 6 自由度的掃描對地圖最佳化之間切換，以減少垂直方向漂移。","Cascaded visual-LiDAR odometry: a semi-direct depth-enhanced monocular VO (direct pose, point matching with propagation, windowed BA; LiDAR depth without interpolation) feeds a CT-ICP-based LiDAR odometry that uses sweep reconstruction to match the camera rate and an adaptive 3-DoF or 6-DoF sweep-to-map optimization.","full_text_reviewed","peer_reviewed_published","supplementary","SDV-LOAM 屬於相機與 LiDAR 串接的里程計，評估全在 KITTI 系列車載資料與自建車載設備的定性結果，沒有室內或施工現場測試。其自適應 3 與 6 自由度最佳化是為了處理地面主導、垂直約束弱的場景，對開闊工地的車載或手推式掃描有參考價值（推論），但缺乏迴圈閉合，長距離點雲的全域一致性需另行處理。",[20,21],"public_benchmark","simulation",[23,24,25,26,27],"Visual module KITTI 00-10 average RTE 0.72 % versus 1.16 % for DEMO and 0.94 % for Huang et al. (Table V)","Full system KITTI 00-10 average 0.47 % versus 0.49 % for CT-ICP+ and 0.52 % for MULLS+ (Table VI)","KITTI online test set 0.60 % translational drift, reported as 8th at the time of writing (Sec. VII)","The adaptive sweep-to-map optimization was applied to six open-source LiDAR odometry systems to reduce vertical drift (Sec. VII ablations)","Code released (GPL-2.0)",[29,30,31,32],"No loop closure or global optimization","Own-rig experiments are qualitative (trajectory overlaid on a map; no ground truth)","Baseline values in Table V come from the original publications because some codes could not be compiled or were not released (Sec. VII)","On the KITTI test set V-LOAM (0.54 %) remains lower than SDV-LOAM (0.60 %) (Table VI)",[34,35],"monocular grayscale camera","3D LiDAR (Velodyne HDL-64E on KITTI; VLP-16 on the authors' rig)",[37,38],"KITTI, KITTI-360 and KITTI-CARLA vehicles","authors' camera-LiDAR rig on a vehicle (qualitative)","semi-direct depth-enhanced visual odometry built on DSO ideas: direct photometric pose estimate, then point matching with propagation, reprojection-based refinement and sliding-window bundle adjustment with marginalization; its pose is the motion prior for a CT-ICP-based LiDAR odometry with adaptive sweep-to-map optimization (Secs. V-VI)","visual tracking points take depth from projected LiDAR points without depth interpolation, plus extra high-gradient points without LiDAR depth; LiDAR sweep-to-map point-to-plane registration against a voxel map (20 points per voxel) switching between 3-DoF and 6-DoF optimization by the ratio of ground to vertical constraints (threshold 0.8) (Secs. V-VI)","sweep reconstruction splits and recombines 10 Hz LiDAR sweeps into 60 Hz segments aligned with the camera frames (Sec. VI; Fig. 5); the LiDAR module is built on CT-ICP, a continuous-time registration method","not described as a separate step; the LiDAR module is based on CT-ICP (inference: intra-sweep motion is handled by its continuous-time model)","no","none","voxel map of LiDAR points for sweep-to-map registration; global point cloud map output (Fig. 8)","camera-LiDAR extrinsic calibration (Autoware calibration toolkit for the authors' rig)","vehicle trajectory and global LiDAR point cloud map","laptop with Intel i7-11700 and 16 GB RAM; visual odometry about 0.06 s per frame; both modules reach about 20 Hz in practice (Sec. VII)","https:\u002F\u002Fgithub.com\u002FZikangYuan\u002FSDV-LOAM","GPL-2.0 (GitHub license metadata)",[52],{"relation":53,"title":54,"doi_or_url":49},"code_release","ZikangYuan\u002FSDV-LOAM (GPL-2.0)",{"id":5,"kind":56,"shortName":7,"title":8,"authors":57,"year":9,"venue":63,"venueType":64,"publisher":65,"volumeIssuePages":66,"doi":67,"arxivId":68,"url":69,"firstPublicDate":70,"publicationStatus":16,"metadataStatus":71,"fulltextStatus":15,"era":10,"classicReason":72,"codeUrl":49,"cluster":11,"topics":73,"mdpi":74,"verification":75,"label":6,"fulltextRoute":76,"versionRead":77,"addedByCensus":78},"method",[58,59,60,61,62],"Zikang Yuan","Qingjie Wang","Ken Cheng","Tianyu Hao","Xin Yang","IEEE Transactions on Pattern Analysis and Machine Intelligence","journal","IEEE","45(9), pp. 11203-11220","10.1109\u002Ftpami.2023.3262817",null,"https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2023.3262817","2023-03-29","metadata_verified","not_applicable",[11],false,"corrected","NTU institutional (Chrome)","IEEE version of record (TPAMI HTML through NTU access, all sections and table images)",true,[80,87,92,98,103,108],{"category":81,"model":82,"canonical":82,"role":83,"dataset":84,"specs":85,"locator":86},"camera","FL3-U3-13E4M-C (text also names FL3-FW-14S3M-C)","method input","SDV-LOAM own platform (qualitative)","grayscale images 1280 x 1040 at 60 Hz","Sec. VII; Fig. 6",{"category":88,"model":89,"canonical":90,"role":83,"dataset":84,"specs":91,"locator":86},"lidar","VLP-16 (introduction says VLP-16E)","Velodyne VLP-16","16-beam LiDAR at 10 Hz",{"category":88,"model":93,"canonical":93,"role":94,"dataset":95,"specs":96,"locator":97},"Velodyne HDL-64E","dataset sensor","KITTI odometry","KITTI LiDAR","Sec. VII",{"category":99,"model":100,"canonical":100,"role":101,"dataset":95,"specs":102,"locator":97},"gnss","KITTI high accuracy GPS\u002FINS","reference or ground truth","ground truth for KITTI",{"category":104,"model":105,"canonical":105,"role":106,"dataset":68,"specs":107,"locator":97},"compute","laptop with Intel i7-11700","compute for runtime","16 GB RAM",{"category":109,"model":110,"canonical":110,"role":94,"dataset":95,"specs":111,"locator":112},"stereo_camera","KITTI color and monochrome stereo cameras (model not named; only the left image is used)","10 Hz; SDV-LOAM takes the LiDAR point cloud and the left image of the stereo camera as input","Sec. VII-A",[],{"totalRows":115,"groupCount":116,"groups":117,"others":469},26,2,[118,290],{"slug":119,"group":120,"sourceId":5,"sourceLabel":6,"table":121,"selfRows":122,"metrics":123,"seqs":128,"entrants":156,"cells":169,"outcomes":281,"locators":284,"hardware":286,"wordings":287,"notes":288},"sdvloam2023-table-v","sdvloam2023:Table V","Table V",13,[124],{"label":125,"unit":126,"statistic":127,"alignment":72},"Relative translational error (RTE)","%","mean",[129,132,134,136,138,140,142,144,146,148,150,152,154],{"dataset":95,"sequence":130,"environment":131},"00","urban, highway and country driving",{"dataset":95,"sequence":133,"environment":131},"01",{"dataset":95,"sequence":135,"environment":131},"02",{"dataset":95,"sequence":137,"environment":131},"03",{"dataset":95,"sequence":139,"environment":131},"04",{"dataset":95,"sequence":141,"environment":131},"05",{"dataset":95,"sequence":143,"environment":131},"06",{"dataset":95,"sequence":145,"environment":131},"07",{"dataset":95,"sequence":147,"environment":131},"08",{"dataset":95,"sequence":149,"environment":131},"09",{"dataset":95,"sequence":151,"environment":131},"10",{"dataset":95,"sequence":153,"environment":131},"00-10 average",{"dataset":95,"sequence":155,"environment":131},"11-21 mean (KITTI test set)",[157,160,163,165,167],{"name":158,"methodId":159,"linkable":78,"proposed":74,"self":74},"DEMO","demo2014",{"name":161,"methodId":162,"linkable":78,"proposed":74,"self":74},"LIMO*","limo2018",{"name":164,"methodId":68,"linkable":74,"proposed":74,"self":74},"Huang et al.",{"name":166,"methodId":68,"linkable":74,"proposed":74,"self":74},"DVL-SLAM",{"name":168,"methodId":5,"linkable":78,"proposed":78,"self":78},"Our VO module",[170,174,177,179,182,185,188,191,194,197,200,203,205,207,209,211,213,215,216,217,218,220,222,224,226,228,230,232,234,236,238,239,241,243,245,247,249,250,252,254,256,258,260,261,262,264,266,268,269,271,273,275,276,278,279],[171,171,171,172,173,171,173,173,171],0,1.05,-1,[171,171,175,176,173,171,173,173,171],1,1.87,[171,171,116,178,173,171,173,173,171],0.93,[171,171,180,181,173,171,173,173,171],3,0.99,[171,171,183,184,173,171,173,173,171],4,1.23,[171,171,186,187,173,171,173,173,171],5,1.04,[171,171,189,190,173,171,173,173,171],6,0.96,[171,171,192,193,173,171,173,173,171],7,1.16,[171,171,195,196,173,171,173,173,171],8,1.24,[171,171,198,199,173,171,173,173,171],9,1.17,[171,171,201,202,173,171,173,173,171],10,1.14,[171,171,204,193,173,171,173,173,171],11,[171,171,206,202,173,171,173,173,171],12,[175,171,171,208,173,171,173,173,171],1.12,[175,171,175,210,173,171,173,173,171],0.91,[175,171,183,212,173,171,173,173,171],0.53,[175,171,204,214,171,171,173,173,171],0.85,[175,171,206,178,173,171,173,173,171],[116,171,171,181,173,171,173,173,171],[116,171,175,176,173,171,173,173,171],[116,171,116,219,173,171,173,173,171],1.38,[116,171,180,221,173,171,173,173,171],0.65,[116,171,183,223,173,171,173,173,171],0.42,[116,171,186,225,173,171,173,173,171],0.72,[116,171,189,227,173,171,173,173,171],0.61,[116,171,192,229,173,171,173,173,171],0.56,[116,171,195,231,173,171,173,173,171],1.27,[116,171,198,233,173,171,173,173,171],1.06,[116,171,201,235,173,171,173,173,171],0.83,[116,171,204,237,173,171,173,173,171],0.94,[180,171,171,178,173,171,173,173,171],[180,171,175,240,173,171,173,173,171],1.47,[180,171,116,242,173,171,173,173,171],1.11,[180,171,180,244,173,171,173,173,171],0.92,[180,171,183,246,173,171,173,173,171],0.67,[180,171,186,248,173,171,173,173,171],0.82,[180,171,189,244,173,171,173,173,171],[180,171,192,251,173,171,173,173,171],1.26,[180,171,195,253,173,171,173,173,171],1.32,[180,171,198,255,173,171,173,173,171],0.66,[180,171,201,257,173,171,173,173,171],0.7,[180,171,204,259,175,171,173,173,171],1.98,[183,171,171,246,173,171,173,173,171],[183,171,175,190,173,171,173,173,171],[183,171,116,263,173,171,173,173,171],0.75,[183,171,180,265,173,171,173,173,171],0.86,[183,171,183,267,173,171,173,173,171],0.77,[183,171,186,255,173,171,173,173,171],[183,171,189,270,173,171,173,173,171],0.44,[183,171,192,272,173,171,173,173,171],0.74,[183,171,195,274,173,171,173,173,171],1.07,[183,171,198,212,173,171,173,173,171],[183,171,201,277,173,171,173,173,171],0.51,[183,171,204,225,173,171,173,173,171],[183,171,206,280,173,171,173,173,171],0.88,[282,283],"average over the sequences reported for LIMO* (00, 01, 04) as printed","as printed; the listed 00-10 values average about 0.98",[285],"Table V; Sec. VII",[],[],[289],"KITTI odometry; relative translational error (%) of LiDAR-assisted depth-enhanced visual odometry; baseline values from the original publications; LIMO* uses semantic information; '-' cells not stored",{"slug":291,"group":292,"sourceId":5,"sourceLabel":6,"table":293,"selfRows":122,"metrics":294,"seqs":296,"entrants":310,"cells":328,"outcomes":462,"locators":463,"hardware":465,"wordings":466,"notes":467},"sdvloam2023-table-vi-kitti-part","sdvloam2023:Table VI (KITTI part)","Table VI (KITTI part)",[295],{"label":125,"unit":126,"statistic":127,"alignment":72},[297,298,299,300,301,302,303,304,305,306,307,308,309],{"dataset":95,"sequence":130,"environment":131},{"dataset":95,"sequence":133,"environment":131},{"dataset":95,"sequence":135,"environment":131},{"dataset":95,"sequence":137,"environment":131},{"dataset":95,"sequence":139,"environment":131},{"dataset":95,"sequence":141,"environment":131},{"dataset":95,"sequence":143,"environment":131},{"dataset":95,"sequence":145,"environment":131},{"dataset":95,"sequence":147,"environment":131},{"dataset":95,"sequence":149,"environment":131},{"dataset":95,"sequence":151,"environment":131},{"dataset":95,"sequence":153,"environment":131},{"dataset":95,"sequence":155,"environment":131},[311,313,315,317,319,321,323,326],{"name":312,"methodId":68,"linkable":74,"proposed":74,"self":74},"A-LOAM+",{"name":314,"methodId":68,"linkable":74,"proposed":74,"self":74},"LeGO-LOAM+",{"name":316,"methodId":68,"linkable":74,"proposed":74,"self":74},"Fast-LOAM+",{"name":318,"methodId":68,"linkable":74,"proposed":74,"self":74},"ISC-LOAM+",{"name":320,"methodId":68,"linkable":74,"proposed":74,"self":74},"MULLS+",{"name":322,"methodId":68,"linkable":74,"proposed":74,"self":74},"CT-ICP+",{"name":324,"methodId":325,"linkable":78,"proposed":74,"self":74},"V-LOAM","vloam2015",{"name":327,"methodId":5,"linkable":78,"proposed":78,"self":78},"Ours (SDV-LOAM)",[329,330,332,334,336,338,340,341,342,344,346,348,350,352,354,356,357,359,361,362,363,365,367,369,371,373,374,376,378,379,380,382,383,384,385,386,387,389,390,391,392,393,395,396,397,398,399,400,402,404,405,407,409,411,413,415,417,419,420,422,423,425,427,428,429,431,433,435,437,438,439,440,441,443,445,446,447,448,450,452,453,454,456,458,459,460],[171,171,171,280,173,171,173,173,171],[171,171,175,331,173,171,173,173,171],2.65,[171,171,116,333,173,171,173,173,171],1.52,[171,171,180,335,173,171,173,173,171],1.15,[171,171,183,337,173,171,173,173,171],1.36,[171,171,186,339,173,171,173,173,171],0.68,[171,171,189,225,173,171,173,173,171],[171,171,192,212,173,171,173,173,171],[171,171,195,343,173,171,173,173,171],1.18,[171,171,198,345,173,171,173,173,171],1.19,[171,171,201,347,173,171,173,173,171],1.57,[171,171,204,349,173,171,173,173,171],1.22,[175,171,171,351,173,171,173,173,171],1.25,[175,171,175,353,173,171,173,173,171],3.35,[175,171,116,355,173,171,173,173,171],1.74,[175,171,180,208,173,171,173,173,171],[175,171,183,358,173,171,173,173,171],1.73,[175,171,186,360,173,171,173,173,171],0.87,[175,171,189,214,173,171,173,173,171],[175,171,192,255,173,171,173,173,171],[175,171,195,364,173,171,173,173,171],1.35,[175,171,198,366,173,171,173,173,171],1.75,[175,171,201,368,173,171,173,173,171],1.96,[175,171,204,370,173,171,173,173,171],1.51,[116,171,171,372,173,171,173,173,171],0.76,[116,171,175,242,173,171,173,173,171],[116,171,116,375,173,171,173,173,171],1.09,[116,171,180,377,173,171,173,173,171],1.67,[116,171,183,267,173,171,173,173,171],[116,171,186,257,173,171,173,173,171],[116,171,189,381,173,171,173,173,171],0.48,[116,171,192,187,173,171,173,173,171],[116,171,195,199,173,171,173,173,171],[116,171,198,272,173,171,173,173,171],[116,171,201,265,173,171,173,173,171],[116,171,204,237,173,171,173,173,171],[180,171,171,388,173,171,173,173,171],1.03,[180,171,175,242,173,171,173,173,171],[180,171,116,184,173,171,173,173,171],[180,171,180,377,173,171,173,173,171],[180,171,183,267,173,171,173,173,171],[180,171,186,394,173,171,173,173,171],0.98,[180,171,189,381,173,171,173,173,171],[180,171,192,187,173,171,173,173,171],[180,171,195,199,173,171,173,173,171],[180,171,198,272,173,171,173,173,171],[180,171,201,265,173,171,173,173,171],[180,171,204,401,173,171,173,173,171],1.01,[183,171,171,403,173,171,173,173,171],0.52,[183,171,175,246,173,171,173,173,171],[183,171,116,406,173,171,173,173,171],0.59,[183,171,180,408,173,171,173,173,171],0.62,[183,171,183,410,173,171,173,173,171],0.46,[183,171,186,412,173,171,173,173,171],0.31,[183,171,189,414,173,171,173,173,171],0.29,[183,171,192,416,173,171,173,173,171],0.34,[183,171,195,418,173,171,173,173,171],0.81,[183,171,198,277,173,171,173,173,171],[183,171,201,421,173,171,173,173,171],0.64,[183,171,204,403,173,171,173,173,171],[186,171,171,424,173,171,173,173,171],0.49,[186,171,175,426,173,171,173,173,171],0.69,[186,171,116,403,173,171,173,173,171],[186,171,180,257,173,171,173,173,171],[186,171,183,430,173,171,173,173,171],0.36,[186,171,186,432,173,171,173,173,171],0.25,[186,171,189,434,173,171,173,173,171],0.28,[186,171,192,436,173,171,173,173,171],0.32,[186,171,195,418,173,171,173,173,171],[186,171,198,381,173,171,173,173,171],[186,171,201,424,173,171,173,173,171],[186,171,204,424,173,171,173,173,171],[189,171,206,442,173,171,173,173,171],0.54,[192,171,171,444,173,171,173,173,171],0.5,[192,171,175,408,173,171,173,173,171],[192,171,116,277,173,171,173,173,171],[192,171,180,229,173,171,173,173,171],[192,171,183,449,173,171,173,173,171],0.37,[192,171,186,451,173,171,173,173,171],0.27,[192,171,189,434,173,171,173,173,171],[192,171,192,436,173,171,173,173,171],[192,171,195,455,173,171,173,173,171],0.79,[192,171,198,457,173,171,173,173,171],0.47,[192,171,201,457,173,171,173,173,171],[192,171,204,457,173,171,173,173,171],[192,171,206,461,173,171,173,173,171],0.6,[],[464],"Table VI; Sec. VII",[],[],[468],"KITTI odometry; relative translational error (%) of visual-LiDAR odometry; '+' = open-source LiDAR odometry modified by the authors to use the SDV-LOAM visual module as motion prior; V-LOAM only has test-set results; '-' cells not stored",[],1790510663439]