[{"data":1,"prerenderedAt":466},["ShallowReactive",2],{"method-gaussianlic2025":3},{"method":4,"reference":54,"equipment":80,"figures":117,"results":118},{"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":28,"platform":32,"estimator":33,"association":34,"timeModel":35,"deskew":36,"loopClosure":37,"globalOptimization":38,"mapRepresentation":39,"prior":40,"outputGeometry":41,"compute":42,"codeUrl":43,"codeLicense":44,"relatedVersions":45},"gaussianlic2025","Lang et al., 2025","Gaussian-LIC","Gaussian-LIC: Real-Time Photo-Realistic SLAM with Gaussian Splatting and LiDAR-Inertial-Camera Fusion",2025,"recent","C09","odometry_with_local_mapping","Gaussian-LIC 以連續時間緊耦合的 LiDAR、慣性與相機里程計（Coco-LIC，每 0.1 秒做一次因子圖最佳化）提供位姿，將著色並降取樣的 LiDAR 點與視覺滑動視窗三角化的 SfM 點一起初始化三維高斯，以補足 LiDAR 未涵蓋的相機視野，並加入天空高斯與曝光仿射模型，以 C++ 與 CUDA 加速達成即時寫實建圖。論文只報告渲染品質（PSNR、SSIM、LPIPS）與執行時間，追蹤比較僅為定性描述，沒有 ATE 或地圖幾何精度數值；作者把提升幾何重建品質列為未來工作。","Tightly coupled continuous-time LiDAR-inertial-camera odometry feeds LiDAR and triangulated points into an online 3DGS map for real-time photorealistic mapping.","full_text_reviewed","peer_reviewed_published","main_body","論文未在營建場域測試；資料為 FAST-LIVO、R3LIVE（固態 LiDAR）與 MCD（旋轉式 LiDAR）序列。",[20],"public_benchmark",[22,23],"Robust tracking where RGB-only and RGB-D 3DGS baselines drift or fail (Sec. IV-B1)","Real-time processing within sensor duration (Sec. IV-B3)",[25,26,27],"No quantitative trajectory accuracy (ATE) or map geometric accuracy is reported; the tracking comparison is qualitative (Sec. IV-B1)","Offline COLMAP plus 3DGS reaches higher PSNR on f0 (32.56 versus 29.89 dB) though not in real time (Table II)","Authors plan to improve odometry with the Gaussian map and to improve geometric reconstruction quality (Sec. V)",[29,30,31],"3D LiDAR","IMU","monocular camera",[],"continuous-time factor-graph sliding-window optimization (Coco-LIC) with point-to-map LiDAR, frame-to-map visual and inertial factors","Coco-LIC odometry with point-to-map LiDAR factors, frame-to-map visual factors and inertial factors (Sec. III-B); a separate VINS-Mono-style visual sliding window tracks Shi-Tomasi corners with KLT only to triangulate SfM points for Gaussian initialization; mapping minimizes an L1 plus D-SSIM re-rendering loss with a per-image exposure affine matrix (Eq. 9)","continuous-time (Coco-LIC)","not described in the full text (v3); poses come from the continuous-time Coco-LIC trajectory optimized every 0.1 s","none reported","no pose graph, global BA or loop closure; the Gaussian map is optimized at each keyframe on K = 100 keyframes sampled from all keyframes to limit forgetting","3D Gaussians initialized from colourized LiDAR points plus triangulated visual SfM points; sky and exposure modelling","none","3D Gaussian map (with sky Gaussians) and rendered images; neither map geometric accuracy nor trajectory accuracy is quantified; authors list improving geometric reconstruction quality as future work","desktop with NVIDIA RTX 3090 (24 GB), Intel Core i7-8700 (3.2 GHz) and 32 GB RAM; C++\u002FCUDA with LibTorch and ROS; on sequence f0 (105 s) tracking and mapping both finish in 105 s (198 s without the acceleration strategies), the only real-time method compared","https:\u002F\u002Fgithub.com\u002FAPRIL-ZJU\u002FGaussian-LIC","not_verified",[46,50],{"relation":47,"title":48,"doi_or_url":49},"preprint","arXiv:2404.06926 (v3, 2025-08-20; latest HTML read)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.06926",{"relation":51,"title":52,"doi_or_url":53},"successor_extension","Gaussian-LIC2: LiDAR-Inertial-Camera Gaussian Splatting SLAM (arXiv:2507.04004; IJRR 2026 acceptance claimed only in the repository README, unverified)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.04004",{"id":5,"kind":55,"shortName":7,"title":8,"authors":56,"year":9,"venue":65,"venueType":66,"publisher":67,"volumeIssuePages":68,"doi":69,"arxivId":70,"url":71,"firstPublicDate":72,"publicationStatus":16,"metadataStatus":73,"fulltextStatus":15,"era":10,"classicReason":74,"codeUrl":43,"cluster":11,"topics":75,"mdpi":76,"verification":77,"label":6,"fulltextRoute":78,"versionRead":79,"addedByCensus":76},"method",[57,58,59,60,61,62,63,64],"Xiaolei Lang","Laijian Li","Chenming Wu","Chen Zhao","Lina Liu","Yong Liu","Jiajun Lv","Xingxing Zuo","2025 IEEE International Conference on Robotics and Automation (ICRA)","conference","IEEE","pp. 8500-8507","10.1109\u002Ficra55743.2025.11128712","2404.06926","https:\u002F\u002Fapi.crossref.org\u002Fworks\u002F10.1109\u002Ficra55743.2025.11128712","2024-04-10","metadata_verified","not_applicable",[11],false,"corrected","arXiv","arXiv v3 (2404.06926v3, 20 Aug 2025; comment 'ICRA 2025'), 8 pages; IEEE ICRA 2025 version of record not accessed",[81,88,91,98,100,104,109,111],{"category":82,"model":83,"canonical":83,"role":84,"dataset":85,"specs":86,"locator":87},"compute","NVIDIA RTX 3090","compute for runtime",null,"24 GB VRAM","Sec. IV-A1",{"category":82,"model":89,"canonical":89,"role":84,"dataset":85,"specs":90,"locator":87},"Intel Core i7-8700","3.2 GHz CPU, 32 GB RAM",{"category":92,"model":93,"canonical":93,"role":94,"dataset":95,"specs":96,"locator":97},"lidar","solid-state LiDAR (model not named)","dataset sensor","FAST-LIVO dataset","used for Gaussian initialization and LiDAR factors","Sec. IV-A2",{"category":92,"model":93,"canonical":93,"role":94,"dataset":99,"specs":96,"locator":97},"R3LIVE dataset",{"category":92,"model":101,"canonical":101,"role":94,"dataset":102,"specs":103,"locator":97},"mechanical spinning LiDAR (model not named)","MCD dataset","large-scale MCD dataset; segments of tuhh_day_02, tuhh_day_03 and tuhh_day_04",{"category":105,"model":106,"canonical":106,"role":94,"dataset":107,"specs":108,"locator":97},"camera","RGB camera (model not named)","FAST-LIVO dataset; R3LIVE dataset","640x512 images; only left images used when stereo is provided",{"category":105,"model":106,"canonical":106,"role":94,"dataset":102,"specs":110,"locator":97},"640x480 images; only left images used when stereo is provided",{"category":112,"model":113,"canonical":113,"role":94,"dataset":114,"specs":115,"locator":116},"imu","IMU (model not named)","FAST-LIVO, R3LIVE and MCD datasets","inertial factors in Coco-LIC","Sec. III-B; Sec. IV-A2",[],{"totalRows":119,"groupCount":120,"groups":121,"others":465},37,4,[122,318,393,424],{"slug":123,"group":124,"sourceId":5,"sourceLabel":6,"table":125,"selfRows":126,"metrics":127,"seqs":138,"entrants":166,"cells":180,"outcomes":312,"locators":313,"hardware":314,"wordings":315,"notes":316},"gaussianlic2025-table-i","gaussianlic2025:Table I","Table I",24,[128,132,136],{"label":129,"unit":130,"statistic":131,"alignment":74},"PSNR (dB)","dB","not_reported",{"label":133,"unit":134,"statistic":135,"alignment":74},"SSIM, Avg. column","unitless","mean",{"label":137,"unit":134,"statistic":135,"alignment":74},"LPIPS, Avg. column",[139,143,145,147,150,152,154,157,159,161,164],{"dataset":140,"sequence":141,"environment":142},"FAST-LIVO","f0 hku2","real-world indoor and outdoor sequences (FAST-LIVO, R3LIVE, MCD)",{"dataset":140,"sequence":144,"environment":142},"f1 LiDAR Degenerate",{"dataset":140,"sequence":146,"environment":142},"f2 Visual Challenge",{"dataset":148,"sequence":149,"environment":142},"R3LIVE","r0 hku_campus_seq_00",{"dataset":148,"sequence":151,"environment":142},"r1 degenerate_seq_00",{"dataset":148,"sequence":153,"environment":142},"r2 degenerate_seq_01",{"dataset":155,"sequence":156,"environment":142},"MCD","m0 tuhh_day_02 segment",{"dataset":155,"sequence":158,"environment":142},"m1 tuhh_day_03 segment",{"dataset":155,"sequence":160,"environment":142},"m2 tuhh_day_04 segment",{"dataset":162,"sequence":163,"environment":142},"FAST-LIVO, R3LIVE and MCD","Avg. average",{"dataset":162,"sequence":165,"environment":142},"average of 9 sequences",[167,169,173,176,178],{"name":168,"methodId":85,"linkable":76,"proposed":76,"self":76},"NeRF-SLAM (train view)",{"name":170,"methodId":171,"linkable":172,"proposed":76,"self":76},"MonoGS (train view)","monogs2024",true,{"name":174,"methodId":175,"linkable":172,"proposed":76,"self":76},"SplaTAM with LiDAR pseudo RGB-D (train view)","splatam2024",{"name":177,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC (train view)",{"name":179,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC (novel view)",[181,185,188,191,194,196,199,202,205,208,211,213,215,217,219,221,223,225,227,229,231,233,235,237,239,241,243,245,247,249,251,253,255,257,259,261,263,265,267,269,271,273,275,277,279,281,283,285,287,289,291,294,296,298,300,302,304,306,308,310],[182,182,182,183,184,182,184,184,182],0,25.56,-1,[182,182,186,187,184,182,184,184,182],1,25.47,[182,182,189,190,184,182,184,184,182],2,17.01,[182,182,192,193,184,182,184,184,182],3,19.53,[182,182,120,195,184,182,184,184,182],21.2,[182,182,197,198,184,182,184,184,182],5,15.02,[182,182,200,201,184,182,184,184,182],6,18.7,[182,182,203,204,184,182,184,184,182],7,18.93,[182,182,206,207,184,182,184,184,182],8,15.4,[182,182,209,210,184,182,184,184,182],9,19.65,[186,182,182,212,184,182,184,184,182],23.58,[186,182,186,214,184,182,184,184,182],23.45,[186,182,189,216,184,182,184,184,182],17.04,[186,182,192,218,184,182,184,184,182],16.19,[186,182,120,220,184,182,184,184,182],15.03,[186,182,197,222,184,182,184,184,182],15.09,[186,182,200,224,184,182,184,184,182],16.41,[186,182,203,226,184,182,184,184,182],13.69,[186,182,206,228,184,182,184,184,182],14.91,[186,182,209,230,184,182,184,184,182],17.27,[189,182,182,232,184,182,184,184,182],25.51,[189,182,186,234,184,182,184,184,182],27.4,[189,182,189,236,184,182,184,184,182],17.84,[189,182,192,238,184,182,184,184,182],17.1,[189,182,120,240,184,182,184,184,182],19.24,[189,182,197,242,184,182,184,184,182],18.3,[189,182,200,244,184,182,184,184,182],13.68,[189,182,203,246,184,182,184,184,182],13.17,[189,182,206,248,184,182,184,184,182],10.01,[189,182,209,250,184,182,184,184,182],18.03,[192,182,182,252,184,182,184,184,182],29.89,[192,182,186,254,184,182,184,184,182],31.28,[192,182,189,256,184,182,184,184,182],23.9,[192,182,192,258,184,182,184,184,182],25.27,[192,182,120,260,184,182,184,184,182],22.47,[192,182,197,262,184,182,184,184,182],23.49,[192,182,200,264,184,182,184,184,182],21.06,[192,182,203,266,184,182,184,184,182],22.87,[192,182,206,268,184,182,184,184,182],20.73,[192,182,209,270,184,182,184,184,182],24.55,[120,182,182,272,184,182,184,184,182],29.28,[120,182,186,274,184,182,184,184,182],30.91,[120,182,189,276,184,182,184,184,182],23.28,[120,182,192,278,184,182,184,184,182],24.52,[120,182,120,280,184,182,184,184,182],22.03,[120,182,197,282,184,182,184,184,182],22.59,[120,182,200,284,184,182,184,184,182],20.19,[120,182,203,286,184,182,184,184,182],22.12,[120,182,206,288,184,182,184,184,182],19.57,[120,182,209,290,184,182,184,184,182],23.83,[182,186,292,293,184,182,184,184,182],10,0.603,[182,189,292,295,184,182,184,184,182],0.405,[186,186,292,297,184,182,184,184,182],0.548,[186,189,292,299,184,182,184,184,182],0.71,[189,186,292,301,184,182,184,184,182],0.592,[189,189,292,303,184,182,184,184,182],0.361,[192,186,292,305,184,182,184,184,182],0.77,[192,189,292,307,184,182,184,184,182],0.231,[120,186,292,309,184,182,184,184,182],0.738,[120,189,292,311,184,182,184,184,182],0.236,[],[125],[],[],[317],"Rendering quality; compared methods mapped with ground-truth poses (MCD) or Gaussian-LIC estimated poses (FAST-LIVO, R3LIVE); FAST-LIVO and R3LIVE use a solid-state LiDAR, MCD a spinning LiDAR",{"slug":319,"group":320,"sourceId":5,"sourceLabel":6,"table":321,"selfRows":206,"metrics":322,"seqs":331,"entrants":335,"cells":347,"outcomes":385,"locators":387,"hardware":388,"wordings":390,"notes":391},"gaussianlic2025-table-ii","gaussianlic2025:Table II","Table II",[323,326,328,330],{"label":324,"unit":325,"statistic":131,"alignment":74},"Tracking time (s)","s",{"label":327,"unit":325,"statistic":131,"alignment":74},"Mapping time (s)",{"label":329,"unit":325,"statistic":131,"alignment":74},"Total time (s)",{"label":129,"unit":130,"statistic":131,"alignment":74},[332],{"dataset":140,"sequence":333,"environment":334},"f0 hku2 (duration 105 s)","FAST-LIVO sequence hku2 (f0); scene type not described in the paper",[336,338,340,342,344,345],{"name":337,"methodId":85,"linkable":76,"proposed":76,"self":76},"NeRF-SLAM",{"name":339,"methodId":171,"linkable":172,"proposed":76,"self":76},"MonoGS",{"name":341,"methodId":175,"linkable":172,"proposed":76,"self":76},"SplaTAM",{"name":343,"methodId":85,"linkable":76,"proposed":76,"self":76},"COLMAP + 3DGS (offline)",{"name":7,"methodId":5,"linkable":172,"proposed":172,"self":172},{"name":346,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC w\u002Fo acceleration",[348,350,352,353,355,357,359,360,362,364,366,368,370,371,372,374,376,377,378,379,380,381,383,384],[182,182,182,349,184,182,182,184,182],105,[182,186,182,351,184,182,182,184,182],288,[182,189,182,351,184,182,182,184,182],[182,192,182,354,184,182,184,184,182],25.23,[186,182,182,356,184,182,182,184,182],181,[186,186,182,358,184,182,182,184,182],387,[186,189,182,358,184,182,182,184,182],[186,192,182,361,184,182,184,184,182],21.42,[189,182,182,363,184,182,182,184,182],954,[189,186,182,365,184,182,182,184,182],1620,[189,189,182,367,184,182,182,184,182],2574,[189,192,182,369,184,182,184,184,182],17.37,[192,182,182,85,182,182,182,184,182],[192,186,182,85,182,182,182,184,182],[192,189,182,373,184,182,182,184,182],6764,[192,192,182,375,184,182,184,184,182],32.56,[120,182,182,349,184,182,182,184,182],[120,186,182,349,184,182,182,184,182],[120,189,182,349,184,182,182,184,182],[120,192,182,252,184,182,184,184,182],[197,182,182,349,184,182,182,184,182],[197,186,182,382,184,182,182,184,182],198,[197,189,182,382,184,182,182,184,182],[197,192,182,252,184,182,184,184,182],[386],"not_applicable (offline batch pipeline)",[321],[389],"NVIDIA RTX 3090 (24 GB) + Intel Core i7-8700 (3.2 GHz), 32 GB RAM",[],[392],"Runtime on sequence f0 (105 s of data) with each method's own estimated poses; real time means finishing within the data duration",{"slug":394,"group":395,"sourceId":5,"sourceLabel":6,"table":396,"selfRows":120,"metrics":397,"seqs":399,"entrants":401,"cells":410,"outcomes":418,"locators":419,"hardware":420,"wordings":421,"notes":422},"gaussianlic2025-table-iii","gaussianlic2025:Table III","Table III",[398],{"label":129,"unit":130,"statistic":131,"alignment":74},[400],{"dataset":140,"sequence":141,"environment":334},[402,404,406,408],{"name":403,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC w\u002Fo exposure modelling",{"name":405,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC w\u002Fo sky modelling",{"name":407,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC w\u002Fo visual SFM points",{"name":409,"methodId":5,"linkable":172,"proposed":172,"self":172},"Gaussian-LIC full",[411,413,415,417],[182,182,182,412,184,182,184,184,182],29.77,[186,182,182,414,184,182,184,184,182],29.76,[189,182,182,416,184,182,184,184,182],29.7,[192,182,182,252,184,182,184,184,182],[],[396],[],[],[423],"Ablation on sequence f0",{"slug":425,"group":426,"sourceId":427,"sourceLabel":428,"table":429,"selfRows":186,"metrics":430,"seqs":434,"entrants":437,"cells":447,"outcomes":456,"locators":458,"hardware":459,"wordings":462,"notes":463},"gslivm2025-iccv-supp-table-6","gslivm2025:ICCV Supp. Table 6","gslivm2025","Xie et al., 2025","ICCV Supp. Table 6",[431],{"label":432,"unit":433,"statistic":131,"alignment":40},"Mapping FPS","fps",[435],{"dataset":436,"sequence":436,"environment":436},"not stated",[438,439,440,443,445],{"name":337,"methodId":85,"linkable":76,"proposed":76,"self":76},{"name":339,"methodId":171,"linkable":172,"proposed":76,"self":76},{"name":441,"methodId":442,"linkable":172,"proposed":76,"self":76},"3DGS","kerbl2023_3dgs",{"name":444,"methodId":5,"linkable":172,"proposed":76,"self":172},"Gaussian-LIC* (preprint result)",{"name":446,"methodId":427,"linkable":172,"proposed":172,"self":76},"Ours",[448,450,452,453,454],[182,182,182,449,184,182,182,184,182],3.1,[186,182,182,451,184,182,182,184,182],5.3,[189,182,182,85,182,182,182,184,182],[192,182,182,292,184,182,186,184,182],[120,182,182,455,184,182,189,184,182],12.56,[457],"not reported ('-' in table)",[429],[436,460,461],"RTX 3090 GPU (as stated in the GS-LIVM supplement text)","Intel Core i9-13900HX (5.50 GHz), 64 GB RAM, NVIDIA RTX 4060 Laptop 8 GB GPU",[],[464],"Mapping FPS; sequence not stated; Gaussian-LIC value copied from its preprint (marked *), which the GS-LIVM text says ran on an RTX 3090; added by second checker from the ICCV supplement",[],1790510665022]