[{"data":1,"prerenderedAt":499},["ShallowReactive",2],{"method-litgs2026":3},{"method":4,"reference":54,"equipment":73,"figures":107,"results":108},{"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":24,"sensors":30,"platform":35,"estimator":37,"association":38,"timeModel":39,"deskew":40,"loopClosure":41,"globalOptimization":42,"mapRepresentation":43,"prior":44,"outputGeometry":45,"compute":46,"codeUrl":47,"codeLicense":48,"relatedVersions":49},"litgs2026","Shi et al., 2026","LIT-GS","LIT-GS: LiDAR-Inertial-Thermal Gaussian Splatting for Illumination-Robust Mapping",2026,"recent","C07","offline_map_refinement","LIT-GS 以熱影像取代易受光照影響的 RGB 光度監督，建立光達、慣性與熱影像的高斯潑濺（Gaussian Splatting）地圖。它以上游 FAST-LIVO2 的具不確定度視覺地圖點作為跨模態錨點建立熱影像與光達的對應，並把加權的光達點到平面殘差放入光束法平差，再於高斯最佳化中加入光達平面正則化，以抑制表面增厚與結構漂移。此方法為離線流程，作者僅報告 Car 場景的訓練時間為 70.6 分鐘。","An offline LiDAR-inertial-thermal Gaussian splatting pipeline that anchors thermal-LiDAR associations on FAST-LIVO2 map points and injects LiDAR plane constraints into both bundle adjustment and Gaussian training.","full_text_reviewed","accepted_author_claim","supplementary","not_reported（私人資料為 Car、Logo、Sundial、Tree-stump、Bushes 五個場景；公開資料為 M2DGR 的 Street-03、Gate-01、Lift-01、Door-02、Hall-02；未見工地）。熱影像與幾何正則化對夜間或低照度隧道檢測可能有參考價值（推論）。",[20],"public_benchmark",[22,23],"Lower EMD than LIV-GaussMap on private and public scenes, notably in plane-dominated scenes (Sec. IV-B2, Table I)","Illumination-robust supervision from thermal imagery in low-light scenes (Sec. I, V)",[25,26,27,28,29],"RGB-LiDAR Gaussian mapping outperforms thermal-LiDAR under good lighting (Sec. V)","Offline training time of 70.6 min for one scene (Sec. IV-C)","Residual thermal-LiDAR extrinsic errors require refinement (Sec. III-B)","Thermal3D-GS has a lower average EMD than LIT-GS on the five private scenes (0.179 vs 0.191; also lower on Sundial, Tree-stump and Bushes), although the text says its EMD values are generally higher (Table I; Sec. IV-B2)","Source and accuracy of the reference point clouds used for EMD are not described (Sec. IV-B2)",[31,32,33,34],"3D LiDAR (Livox Avia, 10 Hz per Fig. 2)","IMU (built into the Livox Avia, 200 Hz per Fig. 2)","visible-light camera (MV-CA013-21UC)","long-wave thermal imager (MV-CI003-GL-N15, 10 Hz per Fig. 2)",[36],"handheld (inferred: the paper uses a modified version of a sensor suite whose repository is named LIV_handhold, and Fig. 1 shows battery, on-board PC and display; the carrying mode is not stated in the text)","upstream FAST-LIVO2 LIV estimate, then offline LiDAR-plane-constrained bundle adjustment (extension of COLMAP-PCD) and differentiable Gaussian optimization","learned thermal feature matching; uncertainty-tagged LIV visual map points as cross-modal anchors; weighted LiDAR point-to-plane residuals in BA and as a splatting regularizer","not_reported (PPS triggers from a microcontroller synchronize LiDAR, IMU and thermal camera; Fig. 1 a3 labels an STM32 board and GPRMC over USB serial)","not_reported","none reported","offline bundle adjustment over poses and triangulated points","3D Gaussians (thermal), initialized from LiDAR voxel map and refined structure","upstream FAST-LIVO2 map points and poses","thermal Gaussian map with rendered images; geometric consistency evaluated with Earth Mover's Distance to reference point clouds (Sec. IV-B2)","offline: training on one scene 70.6 min; rendering about 25.3 ms per 640x512 frame on i7-14700KF with RTX 4080D (Sec. IV-C)",null,"not_verified",[50],{"relation":51,"title":52,"doi_or_url":53},"preprint","LIT-GS arXiv v1 (2026-06-18)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.20424",{"id":5,"kind":55,"shortName":7,"title":8,"authors":56,"year":9,"venue":63,"venueType":51,"publisher":64,"volumeIssuePages":40,"doi":47,"arxivId":65,"url":53,"firstPublicDate":66,"publicationStatus":16,"metadataStatus":67,"fulltextStatus":15,"era":10,"classicReason":68,"codeUrl":47,"cluster":11,"topics":69,"mdpi":70,"verification":71,"label":6,"fulltextRoute":64,"versionRead":72,"addedByCensus":70},"method",[57,58,59,60,61,62],"Shikuan Shi","Chunran Zheng","Jiaming Xu","Tianyong Ye","Tao Yu","Yukang Cui","arXiv (author comment: accepted to IEEE\u002FRSJ IROS 2026)","arXiv","2606.20424","2026-06-18","metadata_verified","not_applicable",[11],false,"corrected","arXiv 2606.20424 v1 (2026-06-18), all 8 pages including Tables I-III; no other version found (arXiv comment and PDF footer claim IROS 2026 acceptance)",[74,80,85,91,96,101,105],{"category":75,"model":76,"canonical":76,"role":77,"dataset":47,"specs":78,"locator":79},"lidar","Livox Avia","method input","built-in IMU; 10 Hz (Fig. 2)","Sec. IV-A; Fig. 1; Fig. 2",{"category":81,"model":82,"canonical":82,"role":77,"dataset":47,"specs":83,"locator":84},"imu","built-in IMU of the Livox Avia","200 Hz (Fig. 2)","Sec. IV-A; Fig. 2",{"category":86,"model":87,"canonical":87,"role":88,"dataset":89,"specs":40,"locator":90},"camera","MV-CA013-21UC visible-light camera","dataset sensor","authors' private thermal dataset (self-collected rig)","Sec. IV-A",{"category":92,"model":93,"canonical":93,"role":77,"dataset":47,"specs":94,"locator":95},"thermal","MV-CI003-GL-N15 long-wave thermal imager","10 Hz (Fig. 2); rendering evaluated at 640 x 512 per frame (Sec. IV-C); native imager resolution not stated","Sec. IV-A; Sec. IV-C; Fig. 2",{"category":97,"model":98,"canonical":98,"role":77,"dataset":47,"specs":99,"locator":100},"other","STM32 microcontroller (PPS synchronization board)","synchronized timers, GPRMC over USB serial (Fig. 1 a3)","Sec. III; Fig. 1 (a3)",{"category":102,"model":103,"canonical":103,"role":104,"dataset":47,"specs":40,"locator":90},"compute","Intel Core i7-14700KF CPU","compute for runtime",{"category":102,"model":106,"canonical":106,"role":104,"dataset":47,"specs":40,"locator":90},"NVIDIA GeForce RTX 4080D GPU",[],{"totalRows":109,"groupCount":110,"groups":111,"others":498},50,2,[112,466],{"slug":113,"group":114,"sourceId":5,"sourceLabel":6,"table":115,"selfRows":116,"metrics":117,"seqs":129,"entrants":158,"cells":167,"outcomes":460,"locators":461,"hardware":462,"wordings":463,"notes":464},"litgs2026-table-i","litgs2026:Table I","Table I",48,[118,122,125,127],{"label":119,"unit":120,"statistic":40,"alignment":121},"PSNR (rendering fidelity, higher is better)","dB (unit not printed)","none",{"label":123,"unit":124,"statistic":40,"alignment":121},"SSIM (higher is better)","unitless",{"label":126,"unit":124,"statistic":40,"alignment":121},"LPIPS (lower is better)",{"label":128,"unit":40,"statistic":40,"alignment":40},"Earth Mover's Distance between reconstructed and reference point clouds (geometric consistency, lower is better)",[130,134,136,138,140,142,144,148,150,152,154,156],{"dataset":131,"sequence":132,"environment":133},"authors' private thermal dataset","Car","outdoor private scenes under varied lighting",{"dataset":131,"sequence":135,"environment":133},"Logo",{"dataset":131,"sequence":137,"environment":133},"Sundial",{"dataset":131,"sequence":139,"environment":133},"Tree-stump",{"dataset":131,"sequence":141,"environment":133},"Bushes",{"dataset":131,"sequence":143,"environment":133},"Avg. (private)",{"dataset":145,"sequence":146,"environment":147},"M2DGR (public)","Street-03","public M2DGR scenes (street, gate, lift, door, hall)",{"dataset":145,"sequence":149,"environment":147},"Gate-01",{"dataset":145,"sequence":151,"environment":147},"Lift-01",{"dataset":145,"sequence":153,"environment":147},"Door-02",{"dataset":145,"sequence":155,"environment":147},"Hall-02",{"dataset":145,"sequence":157,"environment":147},"Avg. (public)",[159,162,164],{"name":160,"methodId":5,"linkable":161,"proposed":161,"self":161},"Ours",true,{"name":163,"methodId":47,"linkable":70,"proposed":70,"self":70},"Thermal3D-GS",{"name":165,"methodId":166,"linkable":161,"proposed":70,"self":70},"LIV-GaussMap","livgaussmap2024",[168,172,175,177,180,183,186,189,192,195,198,201,204,206,208,210,212,214,216,218,220,222,224,226,228,230,232,234,236,238,240,242,244,246,248,250,252,254,256,258,260,262,264,266,268,270,272,274,276,278,280,282,284,286,288,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338,340,342,344,346,348,350,352,354,356,358,359,361,363,365,367,369,371,373,375,377,379,381,383,385,387,389,391,393,395,396,398,400,402,404,406,408,410,412,414,416,418,420,421,423,425,427,429,431,433,434,436,438,440,441,443,444,446,448,450,451,453,455,457,458],[169,169,169,170,171,169,171,171,169],0,26.061,-1,[169,169,173,174,171,169,171,171,169],1,28.04,[169,169,110,176,171,169,171,171,169],25.717,[169,169,178,179,171,169,171,171,169],3,26.565,[169,169,181,182,171,169,171,171,169],4,27.356,[169,169,184,185,171,169,171,171,169],5,26.748,[169,169,187,188,171,169,171,171,169],6,27.127,[169,169,190,191,171,169,171,171,169],7,32.677,[169,169,193,194,171,169,171,171,169],8,35.548,[169,169,196,197,171,169,171,171,169],9,23.057,[169,169,199,200,171,169,171,171,169],10,19.75,[169,169,202,203,171,169,171,171,169],11,27.632,[169,173,169,205,171,169,171,171,169],0.815,[169,173,173,207,171,169,171,171,169],0.833,[169,173,110,209,171,169,171,171,169],0.775,[169,173,178,211,171,169,171,171,169],0.845,[169,173,181,213,171,169,171,171,169],0.793,[169,173,184,215,171,169,171,171,169],0.812,[169,173,187,217,171,169,171,171,169],0.813,[169,173,190,219,171,169,171,171,169],0.903,[169,173,193,221,171,169,171,171,169],0.958,[169,173,196,223,171,169,171,171,169],0.735,[169,173,199,225,171,169,171,171,169],0.714,[169,173,202,227,171,169,171,171,169],0.825,[169,110,169,229,171,169,171,171,169],0.28,[169,110,173,231,171,169,171,171,169],0.323,[169,110,110,233,171,169,171,171,169],0.266,[169,110,178,235,171,169,171,171,169],0.254,[169,110,181,237,171,169,171,171,169],0.232,[169,110,184,239,171,169,171,171,169],0.271,[169,110,187,241,171,169,171,171,169],0.272,[169,110,190,243,171,169,171,171,169],0.372,[169,110,193,245,171,169,171,171,169],0.193,[169,110,196,247,171,169,171,171,169],0.383,[169,110,199,249,171,169,171,171,169],0.356,[169,110,202,251,171,169,171,171,169],0.315,[169,178,169,253,171,169,171,171,169],0.157,[169,178,173,255,171,169,171,171,169],0.235,[169,178,110,257,171,169,171,171,169],0.213,[169,178,178,259,171,169,171,171,169],0.209,[169,178,181,261,171,169,171,171,169],0.14,[169,178,184,263,171,169,171,171,169],0.191,[169,178,187,265,171,169,171,171,169],0.137,[169,178,190,267,171,169,171,171,169],0.077,[169,178,193,269,171,169,171,171,169],0.087,[169,178,196,271,171,169,171,171,169],0.163,[169,178,199,273,171,169,171,171,169],0.098,[169,178,202,275,171,169,171,171,169],0.112,[173,169,169,277,171,169,171,171,169],24.216,[173,169,173,279,171,169,171,171,169],30.152,[173,169,110,281,171,169,171,171,169],26.927,[173,169,178,283,171,169,171,171,169],28.736,[173,169,181,285,171,169,171,171,169],24.243,[173,169,184,287,171,169,171,171,169],26.855,[173,169,187,289,171,169,171,171,169],21.271,[173,169,190,291,171,169,171,171,169],23.042,[173,169,193,293,171,169,171,171,169],28.321,[173,169,196,295,171,169,171,171,169],18.863,[173,169,199,297,171,169,171,171,169],12.838,[173,169,202,299,171,169,171,171,169],20.867,[173,173,169,301,171,169,171,171,169],0.808,[173,173,173,303,171,169,171,171,169],0.819,[173,173,110,305,171,169,171,171,169],0.882,[173,173,178,307,171,169,171,171,169],0.803,[173,173,181,309,171,169,171,171,169],0.824,[173,173,184,311,171,169,171,171,169],0.827,[173,173,187,313,171,169,171,171,169],0.744,[173,173,190,315,171,169,171,171,169],0.778,[173,173,193,317,171,169,171,171,169],0.918,[173,173,196,319,171,169,171,171,169],0.626,[173,173,199,321,171,169,171,171,169],0.796,[173,173,202,323,171,169,171,171,169],0.773,[173,110,169,325,171,169,171,171,169],0.282,[173,110,173,327,171,169,171,171,169],0.22,[173,110,110,329,171,169,171,171,169],0.334,[173,110,178,331,171,169,171,171,169],0.205,[173,110,181,333,171,169,171,171,169],0.39,[173,110,184,335,171,169,171,171,169],0.286,[173,110,187,337,171,169,171,171,169],0.511,[173,110,190,339,171,169,171,171,169],0.496,[173,110,193,341,171,169,171,171,169],0.385,[173,110,196,343,171,169,171,171,169],0.57,[173,110,199,345,171,169,171,171,169],0.514,[173,110,202,347,171,169,171,171,169],0.495,[173,178,169,349,171,169,171,171,169],0.227,[173,178,173,351,171,169,171,171,169],0.297,[173,178,110,353,171,169,171,171,169],0.09,[173,178,178,355,171,169,171,171,169],0.179,[173,178,181,357,171,169,171,171,169],0.103,[173,178,184,355,171,169,171,171,169],[173,178,187,360,171,169,171,171,169],0.148,[173,178,190,362,171,169,171,171,169],0.187,[173,178,193,364,171,169,171,171,169],0.183,[173,178,196,366,171,169,171,171,169],0.231,[173,178,199,368,171,169,171,171,169],0.194,[173,178,202,370,171,169,171,171,169],0.189,[110,169,169,372,171,169,171,171,169],23.569,[110,169,173,374,171,169,171,171,169],31.355,[110,169,110,376,171,169,171,171,169],22.142,[110,169,178,378,171,169,171,171,169],32.204,[110,169,181,380,171,169,171,171,169],30.051,[110,169,184,382,171,169,171,171,169],27.864,[110,169,187,384,171,169,171,171,169],29.222,[110,169,190,386,171,169,171,171,169],25.908,[110,169,193,388,171,169,171,171,169],28.081,[110,169,196,390,171,169,171,171,169],18.34,[110,169,199,392,171,169,171,171,169],17.303,[110,169,202,394,171,169,171,171,169],24.711,[110,173,169,205,171,169,171,171,169],[110,173,173,397,171,169,171,171,169],0.709,[110,173,110,399,171,169,171,171,169],0.65,[110,173,178,401,171,169,171,171,169],0.888,[110,173,181,403,171,169,171,171,169],0.81,[110,173,184,405,171,169,171,171,169],0.774,[110,173,187,407,171,169,171,171,169],0.865,[110,173,190,409,171,169,171,171,169],0.76,[110,173,193,411,171,169,171,171,169],0.893,[110,173,196,413,171,169,171,171,169],0.725,[110,173,199,415,171,169,171,171,169],0.685,[110,173,202,417,171,169,171,171,169],0.786,[110,110,169,419,171,169,171,171,169],0.181,[110,110,173,243,171,169,171,171,169],[110,110,110,422,171,169,171,171,169],0.303,[110,110,178,424,171,169,171,171,169],0.338,[110,110,181,426,171,169,171,171,169],0.302,[110,110,184,428,171,169,171,171,169],0.299,[110,110,187,430,171,169,171,171,169],0.332,[110,110,190,432,171,169,171,171,169],0.417,[110,110,193,419,171,169,171,171,169],[110,110,196,435,171,169,171,171,169],0.322,[110,110,199,437,171,169,171,171,169],0.412,[110,110,202,439,171,169,171,171,169],0.333,[110,178,169,229,171,169,171,171,169],[110,178,173,442,171,169,171,171,169],0.247,[110,178,110,351,171,169,171,171,169],[110,178,178,445,171,169,171,171,169],0.314,[110,178,181,447,171,169,171,171,169],0.298,[110,178,184,449,171,169,171,171,169],0.287,[110,178,187,271,171,169,171,171,169],[110,178,190,452,171,169,171,171,169],0.25,[110,178,193,454,171,169,171,171,169],0.202,[110,178,196,456,171,169,171,171,169],0.121,[110,178,199,247,171,169,171,171,169],[110,178,202,459,171,169,171,171,169],0.224,[],[115],[],[],[465],"Thermal Gaussian-splatting reconstruction quality; private scenes captured with the authors' Livox Avia + thermal rig at different times of day (Fig. 3 caption lists 12:00 p.m. to 6:00 a.m.), public scenes are five M2DGR sequences; EMD reference point cloud source not described; Avg. columns printed by the authors",{"slug":467,"group":468,"sourceId":5,"sourceLabel":6,"table":469,"selfRows":110,"metrics":470,"seqs":478,"entrants":483,"cells":485,"outcomes":490,"locators":491,"hardware":493,"wordings":495,"notes":496},"litgs2026-text-sec-iv-c","litgs2026:Text Sec.IV-C","Text Sec.IV-C",[471,474],{"label":472,"unit":473,"statistic":40,"alignment":121},"training time on the Car scene","min",{"label":475,"unit":476,"statistic":477,"alignment":121},"average rendering time for one 640x512 frame (approximately)","ms","mean",[479,481],{"dataset":131,"sequence":132,"environment":480},"outdoor",{"dataset":131,"sequence":482,"environment":480},"not specified (sentence follows the Car-scene training time)",[484],{"name":7,"methodId":5,"linkable":161,"proposed":161,"self":161},[486,488],[169,169,169,487,171,169,169,171,169],70.6,[169,173,173,489,171,169,169,171,169],25.3,[],[492],"Sec. IV-C",[494],"Intel Core i7-14700KF CPU + NVIDIA GeForce RTX 4080D GPU",[],[497],"Offline training and rendering cost",[],1790510658638]