[{"data":1,"prerenderedAt":495},["ShallowReactive",2],{"method-scancontextpp2022":3},{"method":4,"reference":60,"equipment":80,"figures":125,"results":165},{"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":29,"sensors":35,"platform":38,"estimator":41,"association":42,"timeModel":43,"deskew":44,"loopClosure":45,"globalOptimization":46,"mapRepresentation":47,"prior":48,"outputGeometry":43,"compute":49,"codeUrl":50,"codeLicense":51,"relatedVersions":52},"scancontextpp2022","Kim et al., 2022b","Scan Context++","Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments",2022,"recent","C06","place_recognition_component","Scan Context++ 擴充原 Scan Context，提出極座標的 Polar Context（處理航向旋轉）與直角座標的 Cart Context（處理側向平移）兩種描述子。流程分三段：以檢索鍵（retrieval key）建立 kd-tree 做地點檢索，以對齊鍵（aligning key）做 1 自由度半度量定位，最後以完整描述子比對剔除誤判；並以描述子增強同時應對旋轉與側移。方法假設橫滾與俯仰變化不劇烈。","Scan Context++ adds polar and Cartesian variants of the descriptor with retrieval and aligning sub-keys for three-stage place retrieval, 1-DoF semi-metric alignment and verification, robust to heading and lateral changes.","full_text_reviewed","peer_reviewed_published","main_body","未於工地測試；作者明言方法最適合都市結構環境（Sec. VIII-H3），其失效案例為車載資料中沿「走廊狀場所」行駛及公車等高大物體緊貼感測器（Sec. VIII-F），並非室內走廊實驗。作者另以步行手持光達序列 KA Urban Campus 1 測試，結果優於 M2DP，並認為溫和的橫滾、俯仰與高度擾動可接受；但以 ±5° 至 ±15° 隨機預旋轉模擬時，各方法效能皆明顯下降（Sec. VIII-D、Fig. 20）。營建室內走廊、重複樓層與遮蔽嚴重的材料堆放區是否同樣失效屬推論，尚無直接證據；非商用授權（CC BY-NC-SA 4.0）限制工程部署。",[20,21],"public_benchmark","cross_site",[23,24,25,26,27,28],"Global retrieval without odometry prior; lightweight and easy to add to keyframe pose-graph SLAM (Sec. V-A, VII-F)","Evaluated across KITTI, MulRan, Oxford Radar RobotCar and NAVER LABS data (Sec. VI-A)","Multi-session place recognition with temporal gaps, including a two-month gap on MulRan Sejong (Sec. VII-D)","SC-LeGO-LOAM reduced mean translational ATE from 20.7 m to 3.4 m on KAIST 03 and from 47.7 m to 15.2 m on Riverside 02 relative to LeGO-LOAM (Table IV)","Semi-metric 1-DoF alignment errors of 1.03 deg (A-PC) and 0.84 m (A-CC) on Pangyo (Sec. VII-E)","On a hand-held sequence with mild roll, pitch and height perturbation it outperformed M2DP by a large margin (Sec. VIII-D)",[30,31,32,33,34],"Assumes roll-pitch motions are not severe (abstract); random roll and pitch pre-rotations of +\u002F-5, +\u002F-10 and +\u002F-15 deg caused clear performance losses for PC, CC and M2DP (Sec. VIII-D, Fig. 20)","Failure cases shown on driving data: the vehicle moving along a corridor-like place, and a tall, large object (e.g., a bus) very close to the sensor in both query and map scans (Sec. VIII-F, Fig. 22)","Most powerful in urban environments; indoor and natural environments may need extra channels such as intensity or semantics (Sec. VIII-H3)","Heterogeneous LiDAR setups (e.g. different mounting height between mapper and localizer) declared out of scope; generalization across hardware and mounting left open (Sec. IV-C, VIII-H5)","Each descriptor is natively invariant in only one direction (Sec. VIII-H1); without augmentation PC is weaker under lateral change (Riverside 02 AUC 0.72 vs 0.88 for CC) and CC fails on reverse revisits (KITTI 08 AUC 0.00) (Table VIII, IX; Sec. VII-B)",[36,37],"3D LiDAR","radar (extension discussed)",[39,40],"vehicle (KITTI, MulRan, Oxford Radar RobotCar, NAVER LABS Pangyo)","handheld (KA Urban Campus 1 sequence from LiLi-OM, slowly walking operator)","not_applicable (integration example uses iSAM2 pose graph in SC-LeGO-LOAM, Sec. VII-F)","Polar Context (20 x 60 bins over 0 to 80 m and 360 deg) and Cart Context (40 x 40 bins over -100 to 100 m by -40 to 40 m) with maximum-height bins after 0.5 m voxel downsampling; retrieval key = L1 norm per row (the IROS 2018 version used an L0 occupancy ratio) in a single k-d tree with k = 1 candidate; aligning key (L1 per column) gives the column shift by L2 matching; full-descriptor cosine distance only at that shift for verification; augmentation by +\u002F-2 m lateral root shifts (A-PC) or a double flip (A-CC)","not_applicable","not_reported","provides loop candidates plus 1-DoF initial alignment for ICP in a keyframe pose graph","none (component); demonstrated with iSAM2 in SC-LeGO-LOAM","per-keyframe 2D descriptor database","none; no odometry prior needed for retrieval (Sec. V-A, VII-F)","single C++ header and source pair; on Intel i9-9900 3.10 GHz with 64 GB RAM inside the real-time SLAM, mean 7.36 ms per query on KITTI 00 and 7.31 ms on the 31 km Pangyo sequence, with periodic k-d tree rebuild as the most expensive step; Table V totals: PC 8.3 ms, A-PC 11.5 ms and M2DP 5.8 ms measured in Matlab, SegMatch 796.0 ms copied from its paper, PointNetVLAD 34.0 ms on a GTX 1080 Ti GPU","https:\u002F\u002Fgithub.com\u002Fgisbi-kim\u002Fscancontext","CC BY-NC-SA 4.0 (stated in README; non-commercial)",[53,57],{"relation":54,"title":55,"doi_or_url":56},"conference_version","Scan Context (IROS 2018) - predecessor extended by this paper","10.1109\u002FIROS.2018.8593953",{"relation":58,"title":59,"doi_or_url":50},"code_release","gisbi-kim\u002Fscancontext; SC-LeGO-LOAM integration",{"id":5,"kind":61,"shortName":7,"title":8,"authors":62,"year":9,"venue":66,"venueType":67,"publisher":68,"volumeIssuePages":69,"doi":70,"arxivId":71,"url":72,"firstPublicDate":73,"publicationStatus":16,"metadataStatus":74,"fulltextStatus":15,"era":10,"classicReason":43,"codeUrl":50,"cluster":11,"topics":75,"mdpi":76,"verification":77,"label":6,"fulltextRoute":78,"versionRead":79,"addedByCensus":76},"component",[63,64,65],"Giseop Kim","Sunwook Choi","Ayoung Kim","IEEE Transactions on Robotics","journal","IEEE","38(3):1856-1874","10.1109\u002Ftro.2021.3116424","2109.13494","https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.13494","2021-09-28","metadata_verified",[11],false,"corrected","arXiv","arXiv 2109.13494v1 (28 Sep 2021, T-RO manuscript layout, 19 pages) read in full; IEEE Xplore version of record HTML (document 9610172, T-RO 38(3):1856-1874, published 10 Nov 2021) spot-checked for section structure and key values (7.36 ms, 7.31 ms, i9-9900, KA Urban Campus, 1.03, 0.84 m, SC-LeGO-LOAM, corridor-like); VoR table images not re-checked because IEEE Xplore became temporarily unavailable",[81,88,93,99,104,109,114,121],{"category":82,"model":83,"canonical":83,"role":84,"dataset":85,"specs":86,"locator":87},"lidar","Velodyne HDL-64E","dataset sensor","KITTI odometry","64-ray, full HFOV","Sec. VI-A1; Table III",{"category":82,"model":89,"canonical":89,"role":84,"dataset":90,"specs":91,"locator":92},"Ouster OS1-64","MulRan","64-ray, 290 deg usable HFOV in Table III","Sec. VI-A2; Table III",{"category":82,"model":94,"canonical":95,"role":84,"dataset":96,"specs":97,"locator":98},"Velodyne HDL-32E (two, left and right of the radar)","Velodyne HDL-32E","Oxford Radar RobotCar","32-ray; scans concatenated into one cloud","Sec. VI-A3; Table III",{"category":100,"model":101,"canonical":101,"role":84,"dataset":96,"specs":102,"locator":103},"gnss","INS and GPS (model not stated)","sequences chosen where INS and GPS were available over the whole trajectory","Sec. VI-A3",{"category":82,"model":105,"canonical":105,"role":84,"dataset":106,"specs":107,"locator":108},"not_reported (32-ray LiDAR)","NAVER LABS (Pangyo)","32-ray, full HFOV","Table III; Sec. VII-G",{"category":82,"model":110,"canonical":110,"role":84,"dataset":111,"specs":112,"locator":113},"not_reported (hand-held LiDAR of the LiLi-OM KA Urban Campus 1 sequence)","LiLi-OM KA Urban Campus 1","narrow front horizontal FOV about 70 deg","Sec. VIII-D",{"category":115,"model":116,"canonical":116,"role":117,"dataset":118,"specs":119,"locator":120},"compute","Intel i9-9900 CPU","compute for runtime",null,"3.10 GHz, 64 GB RAM","Sec. VII-G",{"category":115,"model":122,"canonical":122,"role":117,"dataset":118,"specs":123,"locator":124},"GTX 1080 Ti","GPU used for PointNetVLAD timing","Sec. VII-G; Table V",[126,139,149,157],{"refId":5,"refLabel":6,"fig":127,"whatZh":128,"license":129,"licenseUrl":130,"sourceUrl":131,"src":132,"width":133,"height":134,"thumb":135,"thumbWidth":136,"thumbHeight":137,"modified":138},"Fig. 3(a)","樣本點雲上 Polar Context（黃色）與 Cart Context（灰色）的分格方式，紅色箭頭為對齊軸，綠色箭頭為檢索軸。","CC BY-NC-SA 4.0","https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-nc-sa\u002F4.0\u002F","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2109.13494v1\u002Fscd.png","\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-3-a.webp",1149,1155,"\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-3-a.thumb.webp",480,483,"converted to WebP",{"refId":5,"refLabel":6,"fig":140,"whatZh":141,"license":129,"licenseUrl":130,"sourceUrl":142,"src":143,"width":144,"height":145,"thumb":146,"thumbWidth":136,"thumbHeight":147,"modified":148},"Fig. 1(a)","反向重訪造成的旋轉位移：世界座標與感測器座標下的查詢與資料庫掃描比較。","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2109.13494v1\u002Finvexample_rot_compressed.png","\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-1-a.webp",1400,1736,"\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-1-a.thumb.webp",595,"resized to at most 1400 px wide and converted to WebP",{"refId":5,"refLabel":6,"fig":150,"whatZh":151,"license":129,"licenseUrl":130,"sourceUrl":152,"src":153,"width":144,"height":154,"thumb":155,"thumbWidth":136,"thumbHeight":156,"modified":148},"Fig. 6","MulRan、Oxford Radar RobotCar 與 NAVER LABS 各序列軌跡疊合於航照圖。","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2109.13494v1\u002Fairmap.png","\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-6.webp",321,"\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-6.thumb.webp",110,{"refId":5,"refLabel":6,"fig":158,"whatZh":159,"license":129,"licenseUrl":130,"sourceUrl":160,"src":161,"width":144,"height":162,"thumb":163,"thumbWidth":136,"thumbHeight":164,"modified":148},"Fig. 22(b)","失效案例：車輛沿走廊狀場所行駛造成的感知混淆。","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2109.13494v1\u002Ffail_case1.png","\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-22-b.webp",478,"\u002Ffigure-files\u002Fscancontextpp2022\u002Ffig-22-b.thumb.webp",164,{"totalRows":166,"groupCount":167,"groups":168,"others":483},66,6,[169,300,352,417],{"slug":170,"group":171,"sourceId":5,"sourceLabel":6,"table":172,"selfRows":173,"metrics":174,"seqs":184,"entrants":221,"cells":227,"outcomes":294,"locators":295,"hardware":296,"wordings":297,"notes":298},"scancontextpp2022-table-ix","scancontextpp2022:Table IX","Table IX",32,[175,178,180,182],{"label":176,"unit":177,"statistic":44,"alignment":43},"AUC at correctness threshold 8 m","ratio",{"label":179,"unit":177,"statistic":44,"alignment":43},"AUC at correctness threshold 4 m",{"label":181,"unit":177,"statistic":44,"alignment":43},"AUC at correctness threshold 2 m",{"label":183,"unit":177,"statistic":44,"alignment":43},"AUC at correctness threshold 1 m",[185,188,190,192,194,197,199,201,203,206,208,210,212,215,217,219],{"dataset":85,"sequence":186,"environment":187},"KITTI 00 (threshold 8 m)","urban driving",{"dataset":85,"sequence":189,"environment":187},"KITTI 00 (threshold 4 m)",{"dataset":85,"sequence":191,"environment":187},"KITTI 00 (threshold 2 m)",{"dataset":85,"sequence":193,"environment":187},"KITTI 00 (threshold 1 m)",{"dataset":90,"sequence":195,"environment":196},"KAIST 03 (threshold 8 m)","campus",{"dataset":90,"sequence":198,"environment":196},"KAIST 03 (threshold 4 m)",{"dataset":90,"sequence":200,"environment":196},"KAIST 03 (threshold 2 m)",{"dataset":90,"sequence":202,"environment":196},"KAIST 03 (threshold 1 m)",{"dataset":90,"sequence":204,"environment":205},"Riverside 02 (threshold 8 m)","riverside roads with repeated trees",{"dataset":90,"sequence":207,"environment":205},"Riverside 02 (threshold 4 m)",{"dataset":90,"sequence":209,"environment":205},"Riverside 02 (threshold 2 m)",{"dataset":90,"sequence":211,"environment":205},"Riverside 02 (threshold 1 m)",{"dataset":85,"sequence":213,"environment":214},"KITTI 08 (threshold 8 m)","urban driving, reverse revisits only",{"dataset":85,"sequence":216,"environment":214},"KITTI 08 (threshold 4 m)",{"dataset":85,"sequence":218,"environment":214},"KITTI 08 (threshold 2 m)",{"dataset":85,"sequence":220,"environment":214},"KITTI 08 (threshold 1 m)",[222,225],{"name":223,"methodId":5,"linkable":224,"proposed":224,"self":224},"Polar Context (PC)",true,{"name":226,"methodId":5,"linkable":224,"proposed":224,"self":224},"Cart Context (CC)",[228,232,235,238,241,243,245,247,248,251,253,254,257,258,259,260,261,264,267,270,272,273,275,276,277,280,283,286,289,290,292,293],[229,229,229,230,231,229,231,231,229],0,0.84,-1,[229,233,233,234,231,229,231,231,229],1,0.88,[229,236,236,237,231,229,231,231,229],2,0.91,[229,239,239,240,231,229,231,231,229],3,0.94,[233,229,229,242,231,229,231,231,229],0.81,[233,233,233,244,231,229,231,231,229],0.85,[233,236,236,246,231,229,231,231,229],0.87,[233,239,239,234,231,229,231,231,229],[229,229,249,250,231,229,231,231,229],4,0.99,[229,233,252,250,231,229,231,231,229],5,[229,236,167,250,231,229,231,231,229],[229,239,255,256,231,229,231,231,229],7,0.96,[233,229,249,250,231,229,231,231,229],[233,233,252,250,231,229,231,231,229],[233,236,167,250,231,229,231,231,229],[233,239,255,256,231,229,231,231,229],[229,229,262,263,231,229,231,231,229],8,0.72,[229,233,265,266,231,229,231,231,229],9,0.73,[229,236,268,269,231,229,231,231,229],10,0.79,[229,239,271,266,231,229,231,231,229],11,[233,229,262,234,231,229,231,231,229],[233,233,265,274,231,229,231,231,229],0.89,[233,236,268,246,231,229,231,231,229],[233,239,271,242,231,229,231,231,229],[229,229,278,279,231,229,231,231,229],12,0.55,[229,233,281,282,231,229,231,231,229],13,0.46,[229,236,284,285,231,229,231,231,229],14,0.38,[229,239,287,288,231,229,231,231,229],15,0.23,[233,229,278,229,231,229,231,231,229],[233,233,281,291,231,229,231,231,229],0.01,[233,236,284,291,231,229,231,231,229],[233,239,287,229,231,229,231,231,229],[],[172],[],[],[299],"AUC with respect to correctness threshold (baseline 8 m)",{"slug":301,"group":302,"sourceId":5,"sourceLabel":6,"table":303,"selfRows":304,"metrics":305,"seqs":308,"entrants":317,"cells":326,"outcomes":346,"locators":347,"hardware":348,"wordings":349,"notes":350},"scancontextpp2022-table-viii","scancontextpp2022:Table VIII","Table VIII",16,[306],{"label":307,"unit":177,"statistic":44,"alignment":43},"AUC (area under precision-recall curve)",[309,311,313,315],{"dataset":85,"sequence":310,"environment":187},"KITTI 00",{"dataset":90,"sequence":312,"environment":196},"KAIST 03",{"dataset":90,"sequence":314,"environment":205},"Riverside 02",{"dataset":85,"sequence":316,"environment":214},"KITTI 08",[318,320,322,324],{"name":319,"methodId":5,"linkable":224,"proposed":224,"self":224},"Polar Context (PC), retrieval key",{"name":321,"methodId":5,"linkable":224,"proposed":224,"self":224},"Polar Context (PC), full descriptor",{"name":323,"methodId":5,"linkable":224,"proposed":224,"self":224},"Cart Context (CC), retrieval key",{"name":325,"methodId":5,"linkable":224,"proposed":224,"self":224},"Cart Context (CC), full descriptor",[327,328,329,331,333,334,335,336,337,338,340,341,342,343,344,345],[229,229,229,230,231,229,231,231,229],[233,229,229,244,231,229,231,231,229],[236,229,229,330,231,229,231,231,229],0.8,[239,229,229,332,231,229,231,231,229],0.34,[229,229,233,250,231,229,231,231,229],[233,229,233,250,231,229,231,231,229],[236,229,233,250,231,229,231,231,229],[239,229,233,250,231,229,231,231,229],[229,229,236,263,231,229,231,231,229],[233,229,236,339,231,229,231,231,229],0.74,[236,229,236,234,231,229,231,231,229],[239,229,236,230,231,229,231,231,229],[229,229,239,279,231,229,231,231,229],[233,229,239,282,231,229,231,231,229],[236,229,239,229,231,229,231,231,229],[239,229,239,229,231,229,231,231,229],[],[303],[],[],[351],"AUC of precision-recall: retrieval key (k-d tree, k = 1) vs brute-force full descriptor search; 8 m correctness threshold",{"slug":353,"group":354,"sourceId":5,"sourceLabel":6,"table":355,"selfRows":262,"metrics":356,"seqs":369,"entrants":372,"cells":378,"outcomes":411,"locators":412,"hardware":413,"wordings":414,"notes":415},"scancontextpp2022-table-iv","scancontextpp2022:Table IV","Table IV",[357,361,364,367],{"label":358,"unit":359,"statistic":360,"alignment":44},"ATE Trans. mean","m","mean",{"label":362,"unit":359,"statistic":363,"alignment":44},"ATE Trans. max","max",{"label":365,"unit":366,"statistic":360,"alignment":44},"ATE Rot. mean","deg",{"label":368,"unit":366,"statistic":363,"alignment":44},"ATE Rot. max",[370,371],{"dataset":90,"sequence":312,"environment":196},{"dataset":90,"sequence":314,"environment":205},[373,376],{"name":374,"methodId":375,"linkable":224,"proposed":76,"self":76},"LeGO-LOAM","legoloam2018",{"name":377,"methodId":5,"linkable":224,"proposed":224,"self":224},"SC-LeGO-LOAM",[379,381,383,385,387,389,391,393,395,397,399,401,403,405,407,409],[229,229,229,380,231,229,231,231,229],20.7,[229,233,229,382,231,229,231,231,229],42.7,[229,236,229,384,231,229,231,231,229],4.9,[229,239,229,386,231,229,231,231,229],9.9,[233,229,229,388,231,229,231,231,229],3.4,[233,233,229,390,231,229,231,231,229],8.8,[233,236,229,392,231,229,231,231,229],2.2,[233,239,229,394,231,229,231,231,229],8.2,[229,229,233,396,231,229,231,231,229],47.7,[229,233,233,398,231,229,231,231,229],130.5,[229,236,233,400,231,229,231,231,229],6.9,[229,239,233,402,231,229,231,231,229],12.8,[233,229,233,404,231,229,231,231,229],15.2,[233,233,233,406,231,229,231,231,229],50.5,[233,236,233,408,231,229,231,231,229],4.2,[233,239,233,410,231,229,231,231,229],8.7,[],[355],[],[],[416],"ATE (mean \u002F max) of LeGO-LOAM odometry vs Scan Context integrated SC-LeGO-LOAM (iSAM2 pose graph)",{"slug":418,"group":419,"sourceId":5,"sourceLabel":6,"table":420,"selfRows":167,"metrics":421,"seqs":429,"entrants":431,"cells":444,"outcomes":474,"locators":475,"hardware":476,"wordings":480,"notes":481},"scancontextpp2022-table-v","scancontextpp2022:Table V","Table V",[422,425,427],{"label":423,"unit":424,"statistic":44,"alignment":43},"Description time","ms",{"label":426,"unit":424,"statistic":44,"alignment":43},"Retrieval time",{"label":428,"unit":424,"statistic":44,"alignment":43},"Total time",[430],{"dataset":44,"sequence":44,"environment":44},[432,434,436,439,441],{"name":433,"methodId":5,"linkable":224,"proposed":224,"self":224},"Ours (PC)",{"name":435,"methodId":5,"linkable":224,"proposed":224,"self":224},"Ours (A-PC)",{"name":437,"methodId":438,"linkable":224,"proposed":76,"self":76},"M2DP","m2dp2016",{"name":440,"methodId":118,"linkable":76,"proposed":76,"self":76},"SegMatch",{"name":442,"methodId":443,"linkable":224,"proposed":76,"self":76},"PointNetVLAD","pointnetvlad2018",[445,447,449,451,453,454,456,458,460,462,464,466,468,470,472],[229,229,229,446,231,229,229,231,229],1.6,[229,233,229,448,231,229,229,231,229],6.7,[229,236,229,450,231,229,229,231,229],8.3,[233,229,229,452,231,229,229,231,229],4.8,[233,233,229,448,231,229,229,231,229],[233,236,229,455,231,229,229,231,229],11.5,[236,229,229,457,231,229,229,231,229],4.3,[236,233,229,459,231,229,229,231,229],1.5,[236,236,229,461,231,229,229,231,229],5.8,[239,229,229,463,231,229,233,231,229],430.2,[239,233,229,465,231,229,233,231,229],365.8,[239,236,229,467,231,229,233,231,229],796,[249,229,229,469,231,229,236,231,229],33.3,[249,233,229,471,231,229,236,231,229],0.7,[249,236,229,473,231,229,236,231,229],34,[],[420],[477,478,479],"Matlab, CPU (machine not stated for Table V)","values copied from the SegMatch paper","GPU (GTX 1080 Ti)",[],[482],"Time cost per query in ms; ours and M2DP measured in Matlab, SegMatch copied from its paper, PointNetVLAD on GPU",[484,490],{"group":485,"slug":486,"sourceLabel":6,"table":487,"selfRows":236,"datasets":488},"scancontextpp2022:Text Sec.VII-E","scancontextpp2022-text-sec-vii-e","Text Sec.VII-E",[489],"NAVER LABS",{"group":491,"slug":492,"sourceLabel":6,"table":493,"selfRows":236,"datasets":494},"scancontextpp2022:Text Sec.VII-G","scancontextpp2022-text-sec-vii-g","Text Sec.VII-G",[85,489],1790510654804]