[{"data":1,"prerenderedAt":590},["ShallowReactive",2],{"method-srlivo2024":3},{"method":4,"reference":58,"equipment":80,"figures":125,"results":126},{"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":32,"platform":36,"estimator":39,"association":40,"timeModel":41,"deskew":42,"loopClosure":43,"globalOptimization":44,"mapRepresentation":45,"prior":44,"outputGeometry":46,"compute":47,"codeUrl":48,"codeLicense":49,"relatedVersions":50},"srlivo2024","Yuan et al., 2024","SR-LIVO","SR-LIVO: LiDAR-Inertial-Visual Odometry and Mapping With Sweep Reconstruction",2024,"recent","C07","odometry_with_local_mapping","SR-LIVO 以掃描重組（sweep reconstruction）把光達點流重新切段，使每段掃描的結束時間對齊影像擷取時間，讓較可靠的 LIO 直接估計每張影像當下的位姿。視覺模組因此不再負責狀態估計，只最佳化相機內參、外參與時間偏移，並沿用 R3LIVE 的方式為地圖點上色。","Re-segments LiDAR sweeps to end at image timestamps so that LIO alone estimates camera poses, reducing the vision module to camera-parameter refinement and point coloring.","full_text_reviewed","peer_reviewed_published","supplementary","not_reported",[20],"public_benchmark",[22,23,24,25],"Lowest or tied-lowest RMSE ATE on 8 of 9 NTU-VIRAL sequences against R3Live and Fast-LIVO; Fast-LIVO is better on eee_02 (0.18 vs 0.23 m) (Table I)","Higher PSNR and SSIM of the colorized map than R3Live on all 15 test sequences, e.g., eee_01 PSNR 18.92 vs 12.11 and SSIM 0.82 vs 0.37 (Table II)","About 1.6 times faster than R3Live with comparable colored reconstruction on r3live_01 and a better grayscale map on eee_01 (Sec. VI-D to VI-F, Table VI, Figs. 4 to 6)","Rendering with LIO poses at image times gives higher PSNR and SSIM than rendering with the authors' LiDAR-assisted VIO module on all sequences (Table IV)",[27,28,29,30,31],"Image rate must be down-sampled to at most twice the LiDAR rate, otherwise reconstructed sweeps become too sparse for LIO (Sec. V-A)","No loop closure (Sec. VII)","Premise holds only without LiDAR degeneration: R3Live sequences with LiDAR failure modes were deliberately excluded, and long sequences whose colored map exceeded 32 GB RAM were also excluded (Sec. VI)","Trajectory accuracy is evaluated only on NTU-VIRAL because the R3Live sequences have no position ground truth (Sec. VI-A)","The LIO-over-VIO claim is not uniform: Fast-LIVO's VIO output is more accurate than its LIO output on sbs_01 (0.42 vs 0.73 m) (Table III)",[33,34,35],"3D LiDAR (16-channel OS1 gen1 on NTU-VIRAL; LiDAR written 'LiVOX AVAI' on the R3Live data)","IMU (internal IMU of each LiDAR; LiDAR-IMU extrinsics treated as exact, camera-IMU extrinsics optimized online)","camera (left grayscale camera on NTU-VIRAL; camera of the R3Live handheld rig)",[37,38],"UAV","handheld","ESIKF in the LIO module for all state estimation; separate ESIKF in the vision module optimizing camera intrinsics, extrinsics and time offset only","LIO registration is identical to the authors' SR-LIO and is not re-described (Sec. IV); the vision module tracks map points with Lucas-Kanade optical flow and updates only camera parameters, first by PnP reprojection error and then by photometric error, and renders map-point colors with the R3LIVE rendering function (Sec. V-B)","discrete poses; LiDAR sweeps re-segmented so that sweep end times align with image timestamps","LiDAR points motion-compensated with the IMU-integrated pose; camera images undistorted with offline-calibrated distortion parameters (Sec. III-B)","none; listed as future work (Sec. VII)","none","hash voxel map (as in CT-ICP) with RGB-colored points","dense RGB-colored point cloud map (grayscale on NTU-VIRAL) (Sec. VI-E)","Intel Core i7-11700 with 32 GB RAM; 30 to 34 ms per sweep on R3Live sequences and 14 to 17 ms on NTU-VIRAL, versus 49 to 58 ms for R3Live on R3Live sequences, about 1.6 times faster (Sec. VI-D, Table VI); stable real-time on r3live_01 where R3Live was not (Sec. VI-E, Fig. 4)","https:\u002F\u002Fgithub.com\u002FZikangYuan\u002Fsr_livo","GPL-2.0 (LICENSE file checked)",[51,55],{"relation":52,"title":53,"doi_or_url":54},"preprint","SR-LIVO arXiv (2023-12-28)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.16800",{"relation":56,"title":57,"doi_or_url":48},"code_release","ZikangYuan\u002Fsr_livo",{"id":5,"kind":59,"shortName":7,"title":8,"authors":60,"year":9,"venue":66,"venueType":67,"publisher":68,"volumeIssuePages":69,"doi":70,"arxivId":71,"url":54,"firstPublicDate":72,"publicationStatus":16,"metadataStatus":73,"fulltextStatus":15,"era":10,"classicReason":74,"codeUrl":48,"cluster":11,"topics":75,"mdpi":76,"verification":77,"label":6,"fulltextRoute":78,"versionRead":79,"addedByCensus":76},"method",[61,62,63,64,65],"Zikang Yuan","Jie Deng","Ruiye Ming","Fengtian Lang","Xin Yang","IEEE Robotics and Automation Letters","journal","IEEE","9(6): 5110-5117","10.1109\u002Flra.2024.3389415","2312.16800","2023-12-28","metadata_verified","not_applicable",[11],false,"confirmed","NTU institutional (Chrome)","version of record, IEEE RA-L 9(6):5110-5117 (IEEE Xplore HTML text; table values read from the version-of-record PDF pages 5 to 7); arXiv v1 (2023-12-28) also read in full for comparison",[81,88,93,98,103,107,112,114,116,119],{"category":82,"model":83,"canonical":83,"role":84,"dataset":85,"specs":86,"locator":87},"lidar","16-channel OS1 gen1","dataset sensor","NTU-VIRAL","horizontal 16-channel spinning LiDAR with internal IMU","Sec. VI",{"category":89,"model":90,"canonical":90,"role":84,"dataset":85,"specs":91,"locator":92},"imu","internal IMU of the OS1 LiDAR","LiDAR-IMU extrinsics treated as exact","Sec. III-A, Sec. VI",{"category":94,"model":95,"canonical":95,"role":84,"dataset":85,"specs":96,"locator":97},"camera","left camera","grayscale images","Sec. VI, Sec. VI-F",{"category":99,"model":100,"canonical":100,"role":101,"dataset":85,"specs":102,"locator":87},"other","high-accuracy laser tracking (instrument not named)","reference or ground truth","provides position ground truth",{"category":104,"model":105,"canonical":105,"role":84,"dataset":85,"specs":106,"locator":87},"platform","drone","drone-collected dataset",{"category":82,"model":108,"canonical":109,"role":84,"dataset":110,"specs":111,"locator":87},"LiVOX AVAI","Livox Avia","R3Live dataset","with internal IMU; model name as printed, the paper gives no other model designation",{"category":89,"model":113,"canonical":113,"role":84,"dataset":110,"specs":18,"locator":87},"internal IMU of the Livox LiDAR",{"category":94,"model":94,"canonical":94,"role":84,"dataset":110,"specs":115,"locator":87},"RGB images used for colorization",{"category":104,"model":117,"canonical":117,"role":84,"dataset":110,"specs":118,"locator":87},"handheld device","handheld device-collected dataset; no position ground truth",{"category":120,"model":121,"canonical":121,"role":122,"dataset":123,"specs":124,"locator":87},"compute","Intel Core i7-11700","compute for runtime",null,"consumer-level computer, 32 GB RAM",[],{"totalRows":127,"groupCount":128,"groups":129,"others":589},72,4,[130,312,436,533],{"slug":131,"group":132,"sourceId":5,"sourceLabel":6,"table":133,"selfRows":134,"metrics":135,"seqs":141,"entrants":174,"cells":181,"outcomes":305,"locators":307,"hardware":308,"wordings":309,"notes":310},"srlivo2024-table-ii","srlivo2024:Table II","Table II",30,[136,138],{"label":137,"unit":18,"statistic":18,"alignment":18},"PSNR of map colors versus projected image pixels (higher is better)",{"label":139,"unit":140,"statistic":18,"alignment":18},"SSIM of map colors versus projected image pixels (higher is better)","unitless",[142,145,147,149,151,153,155,158,160,162,164,166,168,170,172],{"dataset":110,"sequence":143,"environment":144},"r3live_01","handheld campus and park sequences (hku_campus, hku_park, hkust_campus)",{"dataset":110,"sequence":146,"environment":144},"r3live_02",{"dataset":110,"sequence":148,"environment":144},"r3live_03",{"dataset":110,"sequence":150,"environment":144},"r3live_04",{"dataset":110,"sequence":152,"environment":144},"r3live_05",{"dataset":110,"sequence":154,"environment":144},"r3live_06",{"dataset":85,"sequence":156,"environment":157},"eee_01","drone-collected sequences (grayscale images)",{"dataset":85,"sequence":159,"environment":157},"eee_02",{"dataset":85,"sequence":161,"environment":157},"eee_03",{"dataset":85,"sequence":163,"environment":157},"nya_01",{"dataset":85,"sequence":165,"environment":157},"nya_02",{"dataset":85,"sequence":167,"environment":157},"nya_03",{"dataset":85,"sequence":169,"environment":157},"sbs_01",{"dataset":85,"sequence":171,"environment":157},"sbs_02",{"dataset":85,"sequence":173,"environment":157},"sbs_03",[175,179],{"name":176,"methodId":177,"linkable":178,"proposed":76,"self":76},"R3Live","r3live2022",true,{"name":180,"methodId":5,"linkable":178,"proposed":178,"self":178},"Ours (SR-LIVO)",[182,186,189,191,193,195,197,199,201,204,206,208,209,212,214,216,217,219,221,223,224,227,229,231,232,235,237,239,241,243,245,246,248,251,253,255,256,259,261,263,265,268,270,271,273,276,278,279,281,284,286,288,290,293,295,297,298,301,303,304],[183,183,183,184,185,183,185,185,183],0,16.76,-1,[187,183,183,188,185,183,185,185,183],1,18.02,[183,187,183,190,185,183,185,185,183],0.73,[187,187,183,192,185,183,185,185,183],0.79,[183,183,187,194,185,183,185,185,183],17.31,[187,183,187,196,185,183,185,185,183],17.73,[183,187,187,198,185,183,185,185,183],0.76,[187,187,187,200,185,183,185,185,183],0.78,[183,183,202,203,185,183,185,185,183],2,14.75,[187,183,202,205,185,183,185,185,183],16.15,[183,187,202,207,185,183,185,185,183],0.69,[187,187,202,198,185,183,185,185,183],[183,183,210,211,185,183,185,185,183],3,15.55,[187,183,210,213,185,183,185,185,183],15.91,[183,187,210,215,185,183,185,185,183],0.72,[187,187,210,190,185,183,185,185,183],[183,183,128,218,185,183,185,185,183],15.48,[187,183,128,220,185,183,185,185,183],16.04,[183,187,128,222,185,183,185,185,183],0.71,[187,187,128,215,185,183,185,185,183],[183,183,225,226,185,183,185,185,183],5,14.28,[187,183,225,228,185,183,185,185,183],15.22,[183,187,225,230,185,183,185,185,183],0.68,[187,187,225,190,185,183,185,185,183],[183,183,233,234,185,183,185,185,183],6,12.11,[187,183,233,236,185,183,185,185,183],18.92,[183,187,233,238,185,183,185,185,183],0.37,[187,187,233,240,185,183,185,185,183],0.82,[183,183,242,123,183,183,185,185,183],7,[187,183,242,244,185,183,185,185,183],20.83,[183,187,242,123,183,183,185,185,183],[187,187,242,247,185,183,185,185,183],0.85,[183,183,249,250,185,183,185,185,183],8,17.34,[187,183,249,252,185,183,185,185,183],22.84,[183,187,249,254,185,183,185,185,183],0.51,[187,187,249,240,185,183,185,185,183],[183,183,257,258,185,183,185,185,183],9,14.07,[187,183,257,260,185,183,185,185,183],19.8,[183,187,257,262,185,183,185,185,183],0.49,[187,187,257,264,185,183,185,185,183],0.84,[183,183,266,267,185,183,185,185,183],10,13.5,[187,183,266,269,185,183,185,185,183],17.75,[183,187,266,230,185,183,185,185,183],[187,187,266,272,185,183,185,185,183],0.86,[183,183,274,275,185,183,185,185,183],11,15.53,[187,183,274,277,185,183,185,185,183],18.42,[183,187,274,192,185,183,185,185,183],[187,187,274,280,185,183,185,185,183],0.88,[183,183,282,283,185,183,185,185,183],12,15.87,[187,183,282,285,185,183,185,185,183],20.28,[183,187,282,287,185,183,185,185,183],0.6,[187,187,282,289,185,183,185,185,183],0.83,[183,183,291,292,185,183,185,185,183],13,17.42,[187,183,291,294,185,183,185,185,183],22.45,[183,187,291,296,185,183,185,185,183],0.52,[187,187,291,192,185,183,185,185,183],[183,183,299,300,185,183,185,185,183],14,15.56,[187,183,299,302,185,183,185,185,183],21.83,[183,187,299,254,185,183,185,185,183],[187,187,299,192,185,183,185,185,183],[306],"diverged",[133],[],[],[311],"Rendering quality of the colorized map projected onto each image, following NeRF-style evaluation; higher is better; 'x' = drifted halfway",{"slug":313,"group":314,"sourceId":5,"sourceLabel":6,"table":315,"selfRows":316,"metrics":317,"seqs":322,"entrants":333,"cells":347,"outcomes":430,"locators":431,"hardware":432,"wordings":433,"notes":434},"srlivo2024-table-iii","srlivo2024:Table III","Table III",18,[318],{"label":319,"unit":320,"statistic":321,"alignment":18},"RMSE of ATE","m","RMSE",[323,325,326,327,328,329,330,331,332],{"dataset":85,"sequence":156,"environment":324},"drone-collected sequences; laser-tracking position ground truth",{"dataset":85,"sequence":159,"environment":324},{"dataset":85,"sequence":161,"environment":324},{"dataset":85,"sequence":163,"environment":324},{"dataset":85,"sequence":165,"environment":324},{"dataset":85,"sequence":167,"environment":324},{"dataset":85,"sequence":169,"environment":324},{"dataset":85,"sequence":171,"environment":324},{"dataset":85,"sequence":173,"environment":324},[334,336,338,341,343,345],{"name":335,"methodId":177,"linkable":178,"proposed":76,"self":76},"R3Live (LIO module)",{"name":337,"methodId":177,"linkable":178,"proposed":76,"self":76},"R3Live(V) (LiDAR-assisted VIO module)",{"name":339,"methodId":340,"linkable":178,"proposed":76,"self":76},"Fast-LIVO (LIO)","fastlivo2022",{"name":342,"methodId":340,"linkable":178,"proposed":76,"self":76},"Fast-LIVO(V) (LiDAR-assisted VIO)",{"name":344,"methodId":5,"linkable":178,"proposed":178,"self":178},"Ours (SR-LIVO LIO)",{"name":346,"methodId":5,"linkable":178,"proposed":76,"self":178},"Ours(V) (authors' R3Live-like LiDAR-assisted VIO module, ablation)",[348,350,352,354,356,358,360,361,362,364,366,368,369,371,373,374,376,378,379,381,382,384,385,386,388,390,392,394,395,396,398,399,401,402,403,404,405,407,409,410,412,414,416,417,418,420,422,423,424,425,426,427,428,429],[183,183,183,349,185,183,185,185,183],1.69,[187,183,183,351,185,183,185,185,183],1.71,[202,183,183,353,185,183,185,185,183],0.28,[210,183,183,355,185,183,185,185,183],0.3,[128,183,183,357,185,183,185,185,183],0.21,[225,183,183,359,185,183,185,185,183],0.24,[183,183,187,123,183,183,185,185,183],[187,183,187,123,183,183,185,185,183],[202,183,187,363,185,183,185,185,183],0.18,[210,183,187,365,185,183,185,185,183],0.26,[128,183,187,367,185,183,185,185,183],0.23,[225,183,187,367,185,183,185,185,183],[183,183,202,370,185,183,185,185,183],0.64,[187,183,202,372,185,183,185,185,183],0.81,[202,183,202,365,185,183,185,185,183],[210,183,202,375,185,183,185,185,183],0.27,[128,183,202,377,185,183,185,185,183],0.22,[225,183,202,375,185,183,185,185,183],[183,183,210,380,185,183,185,185,183],0.63,[187,183,210,380,185,183,185,185,183],[202,183,210,383,185,183,185,185,183],0.33,[210,183,210,383,185,183,185,185,183],[128,183,210,363,185,183,185,185,183],[225,183,210,387,185,183,185,185,183],0.19,[183,183,128,389,185,183,185,185,183],0.35,[187,183,128,391,185,183,185,185,183],0.41,[202,183,128,393,185,183,185,185,183],0.29,[210,183,128,355,185,183,185,185,183],[128,183,128,387,185,183,185,185,183],[225,183,128,397,185,183,185,185,183],0.2,[183,183,225,367,185,183,185,185,183],[187,183,225,400,185,183,185,185,183],0.32,[202,183,225,377,185,183,185,185,183],[210,183,225,393,185,183,185,185,183],[128,183,225,397,185,183,185,185,183],[225,183,225,359,185,183,185,185,183],[183,183,233,406,185,183,185,185,183],0.4,[187,183,233,408,185,183,185,185,183],0.7,[202,183,233,190,185,183,185,185,183],[210,183,233,411,185,183,185,185,183],0.42,[128,183,233,413,185,183,185,185,183],0.12,[225,183,233,415,185,183,185,185,183],0.38,[183,183,242,375,185,183,185,185,183],[187,183,242,383,185,183,185,185,183],[202,183,242,419,185,183,185,185,183],0.25,[210,183,242,421,185,183,185,185,183],0.31,[128,183,242,377,185,183,185,185,183],[225,183,242,377,185,183,185,185,183],[183,183,249,357,185,183,185,185,183],[187,183,249,367,185,183,185,185,183],[202,183,249,359,185,183,185,185,183],[210,183,249,375,185,183,185,185,183],[128,183,249,357,185,183,185,185,183],[225,183,249,377,185,183,185,185,183],[306],[315],[],[],[435],"RMSE of ATE of LIO output versus LiDAR-assisted VIO output within each framework on NTU-VIRAL; 'x' = drifted halfway",{"slug":437,"group":438,"sourceId":5,"sourceLabel":6,"table":439,"selfRows":440,"metrics":441,"seqs":445,"entrants":462,"cells":465,"outcomes":526,"locators":527,"hardware":528,"wordings":530,"notes":531},"srlivo2024-table-vi","srlivo2024:Table VI","Table VI",15,[442],{"label":443,"unit":444,"statistic":18,"alignment":18},"Time consumption per sweep, Total","ms",[446,448,449,450,451,452,453,454,455,456,457,458,459,460,461],{"dataset":110,"sequence":143,"environment":447},"",{"dataset":110,"sequence":146,"environment":447},{"dataset":110,"sequence":148,"environment":447},{"dataset":110,"sequence":150,"environment":447},{"dataset":110,"sequence":152,"environment":447},{"dataset":110,"sequence":154,"environment":447},{"dataset":85,"sequence":156,"environment":447},{"dataset":85,"sequence":159,"environment":447},{"dataset":85,"sequence":161,"environment":447},{"dataset":85,"sequence":163,"environment":447},{"dataset":85,"sequence":165,"environment":447},{"dataset":85,"sequence":167,"environment":447},{"dataset":85,"sequence":169,"environment":447},{"dataset":85,"sequence":171,"environment":447},{"dataset":85,"sequence":173,"environment":447},[463,464],{"name":176,"methodId":177,"linkable":178,"proposed":76,"self":76},{"name":180,"methodId":5,"linkable":178,"proposed":178,"self":178},[466,468,470,472,474,476,478,480,482,484,486,488,490,492,494,496,498,500,502,504,506,508,510,512,514,516,518,520,522,524],[183,183,183,467,185,183,183,185,183],50.57,[187,183,183,469,185,183,183,185,183],30.53,[183,183,187,471,185,183,183,185,183],49.95,[187,183,187,473,185,183,183,185,183],30.78,[183,183,202,475,185,183,183,185,183],49.27,[187,183,202,477,185,183,183,185,183],31.06,[183,183,210,479,185,183,183,185,183],56.14,[187,183,210,481,185,183,183,185,183],30.87,[183,183,128,483,185,183,183,185,183],54.2,[187,183,128,485,185,183,183,185,183],31.42,[183,183,225,487,185,183,183,185,183],57.43,[187,183,225,489,185,183,183,185,183],33.88,[183,183,233,491,185,183,183,185,183],31.02,[187,183,233,493,185,183,183,185,183],16.58,[183,183,242,495,185,183,183,185,183],24.6,[187,183,242,497,185,183,183,185,183],15.1,[183,183,249,499,185,183,183,185,183],23.29,[187,183,249,501,185,183,183,185,183],16.93,[183,183,257,503,185,183,183,185,183],19.67,[187,183,257,505,185,183,183,185,183],13.96,[183,183,266,507,185,183,183,185,183],19.87,[187,183,266,509,185,183,183,185,183],16.49,[183,183,274,511,185,183,183,185,183],20.34,[187,183,274,513,185,183,183,185,183],15.24,[183,183,282,515,185,183,183,185,183],21.31,[187,183,282,517,185,183,183,185,183],14.66,[183,183,291,519,185,183,183,185,183],23.36,[187,183,291,521,185,183,183,185,183],14.79,[183,183,299,523,185,183,183,185,183],22.92,[187,183,299,525,185,183,183,185,183],15.35,[],[439],[529],"Intel Core i7-11700, 32 GB RAM",[],[532],"Total time for handling a sweep as printed in Table VI (per-module Vision and LiDAR columns not extracted); the R3Live totals do not equal the sum of its two module columns, while SR-LIVO totals do",{"slug":534,"group":535,"sourceId":5,"sourceLabel":6,"table":536,"selfRows":257,"metrics":537,"seqs":539,"entrants":549,"cells":554,"outcomes":583,"locators":584,"hardware":585,"wordings":586,"notes":587},"srlivo2024-table-i","srlivo2024:Table I","Table I",[538],{"label":319,"unit":320,"statistic":321,"alignment":18},[540,541,542,543,544,545,546,547,548],{"dataset":85,"sequence":156,"environment":324},{"dataset":85,"sequence":159,"environment":324},{"dataset":85,"sequence":161,"environment":324},{"dataset":85,"sequence":163,"environment":324},{"dataset":85,"sequence":165,"environment":324},{"dataset":85,"sequence":167,"environment":324},{"dataset":85,"sequence":169,"environment":324},{"dataset":85,"sequence":171,"environment":324},{"dataset":85,"sequence":173,"environment":324},[550,551,553],{"name":176,"methodId":177,"linkable":178,"proposed":76,"self":76},{"name":552,"methodId":340,"linkable":178,"proposed":76,"self":76},"Fast-LIVO",{"name":180,"methodId":5,"linkable":178,"proposed":178,"self":178},[555,556,557,558,559,560,561,562,563,564,565,567,568,569,570,571,572,573,574,575,576,577,578,579,580,581,582],[183,183,183,349,185,183,185,185,183],[187,183,183,353,185,183,185,185,183],[202,183,183,357,185,183,185,185,183],[183,183,187,123,183,183,185,185,183],[187,183,187,363,185,183,185,185,183],[202,183,187,367,185,183,185,185,183],[183,183,202,370,185,183,185,185,183],[187,183,202,365,185,183,185,185,183],[202,183,202,377,185,183,185,185,183],[183,183,210,380,185,183,185,185,183],[187,183,210,566,185,183,185,185,183],0.34,[202,183,210,363,185,183,185,185,183],[183,183,128,389,185,183,185,185,183],[187,183,128,393,185,183,185,185,183],[202,183,128,387,185,183,185,185,183],[183,183,225,367,185,183,185,185,183],[187,183,225,393,185,183,185,185,183],[202,183,225,397,185,183,185,185,183],[183,183,233,406,185,183,185,185,183],[187,183,233,190,185,183,185,185,183],[202,183,233,413,185,183,185,185,183],[183,183,242,375,185,183,185,185,183],[187,183,242,419,185,183,185,185,183],[202,183,242,377,185,183,185,185,183],[183,183,249,357,185,183,185,185,183],[187,183,249,359,185,183,185,185,183],[202,183,249,357,185,183,185,185,183],[306],[536],[],[],[588],"RMSE of ATE on NTU-VIRAL; baselines rerun from the authors' source code; 'x' = system drifted halfway through the run",[],1790510656505]