第 8 卷第 6 期 计算机辅助设计与图形学学报 Vol. 8 No.6 016 年 6 月 Journal of Computer-Aided Design & Computer Graphics Jun. 016 基于单目视觉的同时定位与地图构建方法综述 刘浩敏 1), 章国锋 1,)* 1), 鲍虎军 1) ( 浙江大学 CAD&CG 国家重点实验室杭州 310058) ) ( 浙江大学工业信息物理融合系统协同创新中心杭州 310058) (zhangguofeng@cad.zju.edu.cn) : 增强现实是一种在现实场景中无缝地融入虚拟物体或信息的技术, 能够比传统的文字 图像和视频等方式更高效 直观地呈现信息, 有着非常广泛的应用. 同时定位与地图构建作为增强现实的关键基础技术, 可以用来在未知环境中定位自身方位并同时构建环境三维地图, 从而保证叠加的虚拟物体与现实场景在几何上的一致性. 文中首先简述基于视觉的同时定位与地图构建的基本原理 ; 然后介绍几个代表性的基于单目视觉的同时定位与地图构建方法并做深入分析和比较 ; 最后讨论近年来研究热点和发展趋势, 并做总结和展望. : 增强现实 ; 同时定位与地图构建 ; 运动推断结构 ; 多视图几何 ; 摄像机跟踪 :TP391.41 A Survey of Monocular Simultaneous Localization and Mapping Liu Haomin 1), Zhang Guofeng 1,)*, and Bao Hujun 1) 1) (State Key Laboratory of CAD&CG, Zhejiang University, Hangzhou 310058) ) (Collaborative Innovation Center for industrial Cyber-Physical System, Zhejiang University, Hangzhou 310058) Abstract: Augmented reality (AR) is a technique that allows to seamlessly composite virtual objects or information into real scene. Compared to traditional tet, images and videos, AR is a more effective and intuitive way for information presentation and has wide applications. Simultaneous localization and mapping (SLAM) is a key fundamental technique for augmented reality, which provides the ability of self-localization in an unknown environment and mapping the 3D environment simultaneously. The localization and mapping enables fusion of virtual objects and real scenes in a geometrically consistent way. In this paper, we describe the basic principles of Visual SLAM, and introduce some state-of-the-art monocular SLAM methods with deep analysis and comparison. Finally, we discuss some research tendency in recent years and make conclusions. Key words: augmented reality; simultaneous localization and mapping; structure-from-motion; multi-view geometry; camera tracking, ( )().,,., : 016-04-30; : 016-05-11. 基金项目 : (01BAH35B0); (613011, 617048); (015XZZX005-05); (0145). 刘浩敏 (1987),,, ; 章国锋 (1981),,,,,, ; 鲍虎军 (1966),,,,, CCF,.
856 计算机辅助设计与图形学学报第 8 卷,,,..,,., GPS, ; ;. (visual simultaneous localization and mapping, V-SLAM) ( ),., /, / V-SLAM,. (simultaneous localization and mapping, SLAM) [1-4],., (structurefrom-motion, SFM) [5]. SFM, SFM, V-SLAM. SLAM,, SLAM [3-4,6-9]. SLAM, SLAM, SLAM ; ( Durrant- Whyte [3-4] 006 SLAM, 10 ), SLAM., SLAM, 3 V-SLAM,., V-SLAM,.,, [5], 1. V-SLAM V-SLAM,. V-SLAM C 1 C m, X1 X n. C i, 3 3 Ri p i. X j R i, i pi C T ij ij ij i j i ( X, Y, Z ) R ( X p ) (1) h ( f X / Z c, f Y / Z c ) () ij ij ij y ij ij y, f, f y, y, ( c, cy ),. (1)(), h ij Ci X j, h hc (, X ) (3) ij i j V-SLAM ( 1, X 1 11, 1, 31 ). m arg min hc (, X ) ˆ (4), 1 1 C1Cm X1Xn i j n i j ij C 1 C m, X1 X n, X j Ci h ij ij. T 1 ij ~ N( ˆ ij, ij ), e e e. (4)(bundle adjustment, BA) [10], ij T 1 V-SLAM 的基本原理 V-SLAM 1
第 6 期, 等 : 857. V-SLAM,.,,.. V-SLAM IMU (). SLAM VIN(visual-aided inertial navigation) VI-SLAM (visual-inertial SLAM). ( Ci, Ci 1) IMU Z i { z 1 z n i }, VI-SLAM [11-13] arg min hc (, X ) ˆ C1Cm, X1Xn i1 j1 m1 i1 m n f( C, Z ) C i i i1 i i j ij ij (5) (4), VI-SLAM, f ( Ci, Zi) Z i C i, i. (Continuous Time System) [14] (Preintegration) [15]., VI-SLAM v i IMU b i, Ci ( Ri, pi, vi, b i)., GPS p G i,. SLAM. SLAM MonoSLAM [16] MSCKF [17]. MonoSLAM Davison SLAM. MonoSLAM t t C t X 1 X n, ( a );,,,, ( b );, Ct., ( c a. m arg min hc (, X ) ˆ C1Cm, X1Xn i1 j1 n i j ij m1 m1 G f( Ci, Zi) Ci1 i ˆ i i i i1 i1 GPS ij p p (6) G G i i i p ~ N( p ˆ, Λ ). b. 代表性单目 V-SLAM 系统, V-SLAM 3 : BA V-SLAM. V-SLAM,..1 基于滤波器的 V-SLAM V-SLAM : t, t ~ N ( ˆ t, P t), ˆt, Pt. c. MonoSLAM [16]
858 计算机辅助设计与图形学学报第 8 卷 ). MonoSLAM Shi-Tomasi [18], (active search) [19]. MonoSLAM (etended Kalman filter, EKF). (Prediction), Ct f( Ct1, avt, a t) (7) a v a, av ~ N(0, Γ v), a ~ N(0, Γ ), t. (Update), ˆ hc (, X ) n (8) j t j j, ˆ j X j, n j ~ N(0, j ). (5), MonoSLAM t, { C 1 Ct 1 },. EKF : 1) ( V-SLAM ), EKF, Levenberg-Marquardt [0],., ;,,. ), On ( 3 ),. EKF, Mourikis [17] 007 MSCKF. MSCKF VI-SLAM,, IMU [14] ;, MSCKF l C { Ct l 1 Ct}. C C i, C C i,., MSCKF (Marginalization), C i X j 3 { Ct l 1Ct}, On ( ) 3 Onl ( ). l n,,,.. 基于关键帧 BA 的 V-SLAM PTAM SFM, BA V-SLAM, Klein [1] 007, 009 iphone 3G []. PTAM : (Tracking) (Mapping). ( 3a ) ( 3b ), (4)( BA); BA,, C,. t a. b. 3 PTAM [1],, FAST [3], arg min wj h( Ct, X j) ˆ j (9) j Ct n j1 w Tukey [4], j (Outliers). (Inliers) (),, [5],,., C,, t
第 6 期, 等 : 859. (Epipolar Line) [5], (Triangulation) [5], BA,. PTAM V-SLAM, V-SLAM PTAM. Mur-Artal [6] 015 ORB-SLAM V-SLAM. ORB-SLAM PTAM,, 4 : 1) ORB-SLAM ORB [7], ORB [8], PTAM.,,.. ) ORB- SLAM,. [8] (), (Pose Graph). ( 4a ), i ; ( 4b ), ij ; go [9], ( 4c, 4d ),, (, ) 1 m ij ij 1 T 1 1 ij i j ij ij i j argmin ( ) ( ) (10), Σij ij (), a. () b. () c. d. 4 ORB-SLAM [6]. BA,, ORB-SLAM. BA ( 4 ). 3) PTAM,,. (Paralla),. ORB-SLAM. 4) PTAM,. ORB-SLAM, ;, BA. CAD&CG 013 RDSLAM [30], PTAM /, SIFT [31],,., RDSLAM RANSAC [3], RANSAC,,., V-SLAM..3 基于直接跟踪的 V-SLAM BA V-SLAM, ( )., (Direct Tracking),,. V-SLAM DTAM [33] LSD-SLAM [34]. DTAM Newcombe 011 V-SLAM, ( 5a )., DTAM AR ( 5b ),., DTAM
860 计算机辅助设计与图形学学报第 8 卷 C t C v, Cv, C v C t tv, arg min r(, D v ( ), tv ) (11) tv v, r(), r(, Dv, tv) Iv It( (, Dv( ), tv) (1) Iv D v C v, v, (, D v, tv ) v t. I g( ) e r (15) Huber, /, if 1 /, otherwise. DTAM (16) arg min R(, Dr) C(, Dr) d (17) Dr (total variation, TV) [36]. DTAM, DTAM,,, GPU,. Engel [37] 013 (visual odometry, VO), V-SLAM LSD-SLAM [34],. DTAM, LSD-SLAM ( 6a ),,. a. 5 DTAM [33], r D r ( 5c ). (Inverse Depth) [35]. 5d, DTAM M NS, M N, S. DTAM r m N() r, r d D r (Voel) C 1 (, d ) (,, mr ) 1 N() r r d (13) mn( r) r() (1). arg min C(, d), DTAM (Regularization Term) R(, d) g D r ( ) (14), g, d b. 6 LSD-SLAM [34] LSD-SLAM, k I k D k V k. N( Dk, Vk( )). LSD-SLAM t k tk, arg min r Dk tk k r Dk tk (, ( ), ) (, ( ), ) (18), k ; r() (1); (, ) r(),,
第 6 期, 等 : 861 r(, Dk, tk) I r(, Dk, tk) Vk Dk (19) Huber, Outliers. LSD-SLAM (), It Ik, d, EKF D k V k V d D k Dk Vk (0) Vk Vk Vk ORB-SLAM, LSD-SLAM,. k i [38] { k j }, ( k, k ) i j ij (10). ij.4 分析和比较 V-SLAM 1., MSCKF VI-SLAM V-SLAM,, V-SLAM, V-SLAM... MonoSLAM V-SLAM,, PTAM BA V-SLAM [39]. MSCKF,, IMU,. BA PTAM, RDSLAM ORB-SLAM SIFT ORB, PTAM BA, PTAM ; ORB-SLAM,. DTAM LSD-SLAM, ORB-SLAM. [6] TUM RGB-D [40], LSD-SLAM ORB-SLAM 5~10.. MonoSLAM On ( 3 ), n,. 3 MSCKF Onl ( ), n, l. MSCKF, l,,. PTAM ORB-SLAM,. RDSLAM PTAM ORB-SLAM, SIFT. DTAM LSD-SLAM,,,,.. MonoSLAM On ( 3 ),. MSCKF,. PTAM RDSLAM 表 1 各类单目 V-SLAM 系统比较 MonoSLAM MSCKF PTAM ORB-SLAM RDSLAM DTAM LSD-SLAM
86 计算机辅助设计与图形学学报第 8 卷 BA : PTAM ; RDSLAM BA, KDTree [41] PTAM,,. ORB-SLAM LSD-SLAM BA,. DTAM,.. ( ) V-SLAM., DTAM LSD-SLAM,. MSCKF VI-SLAM,, IMU,..,,. MonoSLAM EKF,. PTAM,, MonoSLAM.,,, MonoSLAM. PTAM, DTAM [4],. MSCKF, RDSLAM ORB- SLAM (MSCKF RDSLAM SIFT, ORB-SLAM ORB ), (RDSLAM KDTree [41], ORB-SLAM [43] ), (RDSLAM ). LSD-SLAM,,. PTAM, RDSLAM ORB-SLAM,, RDSLAM ORB-SLAM PTAM. MSCKF IMU,,. LSD- SLAM DTAM,,,..,. MonoSLAM, MSCKF. PTAM,,,.,,. RDSLAM, ORB-SLAM LSD-SLAM (RDSLAM KDTree [41], ORB-SLAM [43], LSD-SLAM FAB-MAP [38] ),, PTAM..,. MonoSLAM,. MSCKF,, MSCKF,, IMU,. PTAM RDSLAM,,,. ORB-SLAM, DTAM LSD-SLAM,. ORB-SLAM,, DTAM LSD-SLAM.. Mono-SLAM, MSCKF, PTAM, ORB-SLAM, DTAM, LSD-SLAM,. RANSAC [3] ( Huber, Tukey ),
第 6 期, 等 : 863, Outliers ;,,. MSCKF IMU,. RDSLAM,, RANSAC,,.,, RDSLAM.. MonoSLAM, ( ),. RDSLAM, SIFT, BA ;, Outliers,. MSCKF, PTAM DTAM. ORB-SLAM LSD-SLAM,.. Klein [5] 008, 7a. LSD-SLAM. 7b,. SFM VO [46-47]. Concha [48] ( 7c ), (Superpiel). Concha [48-49] PTAM LSD-SLAM,., V-SLAM. ( 7d ), [50] [51]. a. [5] b. [37] 3 近年研究热点与发展趋势, V-SLAM.,. 3.1 缓解特征依赖 V-SLAM. (.3 ), /,.,,, [44],. Forster [45] (semi-direct VO, SVO ),,. V-SLAM, (),, c. [48] d. [50] 7 3. 稠密三维重建 V-SLAM., [5], [53]. 010 RGB-D Kinect,.,. Newcombe [54] RGB-D SLAM KinectFusion. KinectFusion TSDF(truncated signed distance function) [55],,. TSDF, Marching Cube [56] TSDF 0, (Ray Cast) TSDF
864 计算机辅助设计与图形学学报第 8 卷 0., KinectFusion, ICP [57],.,, TSDF. KinectFusion : 1) ; ) ICP ; 3)., Whelan : [58],, ; [59] ICP, FOVIS [60] RGB-D [61] 3,, ; [6] DBoW [43], [63], ; [64](Surfel) [65],,,. V-SLAM,., AR,, V-SLAM RGB-D SLAM..3 DTAM LSD-SLAM V- SLAM. Schöps [44] LSD-SLAM,, AR ( 8a ). Pizzoli [66] REMODE,. Tanskanen [67] Kolev [68] ( 8b ), IMU. Pradeep [69] MonoFusion. Ondruska [70] MobileFusion, ( 8c ). Schöps [71] Project Tango, ( 8d ).,,., a. AR [44] b. [67] c. MobileFusion [70] d. [71] 8 (, I I ),. I I, (Patch) NCC(normalized cross correlation)., [67-68],,. MonoFusion [69] PatchMatch [7],.. DTAM REMODE ((14)),. REMODE [73], Inlier,.
第 6 期, 等 : 865 Kolev [68] [65],, Outliers. MonoFusion, MobileFusion Schöps [71] Project Tango TSDF [55] Outliers., GPU. AR, GPU., CPU [34,74],. 3.3 多传感器融合 : V-SLAM ; IMU ; SLAM,.,,. IMU,,. 1,,.,. t, C { C1, C,, C } t, 30 /s, 30,.,,., l { Ct l 1 Ct}, l.,,. SLAM ( l 1).,. MSCKF l. Li [75] MSCKF.0. MSCKF (Observability Analysis), (Linearization Point), (Yaw, ),., MSCKF.0 First-Estimate Jacobian [76],,. [77-78] (). [11-13,79]. EKF, SLAM, EKF,., [11-1] [15] IMU. [79] isam,. [13],. 3.4 其他实际问题 V-SLAM,., V-SLAM, BA V-SLAM (. ),.,,,, ;,,,, V-SLAM. [80-8],,., (Rolling Shutter),,., [83-86].,,, RANSAC, [30,87] IMU. 4 结语,
866 计算机辅助设计与图形学学报第 8 卷. SLAM,., SLAM,., CPU, SLAM. SLAM,,.,, SLAM,. 参考文献 (References): [1] Smith R C, Cheeseman P. On the representation and estimation of spatial uncertainty[j]. International Journal of Robotics Research, 1986, 5(4):56-68 [] Smith R, Self M, Cheeseman P. Estimating uncertain spatial relationships in robotics[m] //Autonomous Robot Vehicles. New York: Springer, 1990: 167-193 [3] Durrant-Whyte H, Bailey T. Simultaneous localization and mapping: Part I[J]. IEEE Robotics & Automation Magazine, 006, 13(): 99-110 [4] Bailey T, Durrant-Whyte H. Simultaneous localization and mapping(slam): Part II[J]. IEEE Robotics & Automation Magazine, 006, 13(3): 108-117 [5] Hartley R, Zisserman A. Multiple view geometry in computer vision[m]. Cambridge: Cambridge University Press, 004 [6] Aulinas J, Petillot Y R, Salvi J, et al. The SLAM problem: a survey[j]. CCIA, 008, 184(1): 363-371 [7] Ros G, Sappa A, Ponsa D, et al. Visual SLAM for driverless cars: a brief survey[c] //Proceedings of IEEE Workshop on Navigation, Perception, Accurate Positioning and Mapping for Intelligent Vehicles. Los Alamitos: IEEE Computer Society Press, 01: Article No.3 [8] He JunueLi Zhanming. Survey of vision-based approach to simultaneous localization and mapping[j]. Application Research of Computer, 010, 7(8): 839-843(in Chinese) (,. [J]., 010, 7(8): 839-843) [9] Liang Mingjie, Min Huaqing, Luo Ronghua. Graph-based SLAM: a survey[j]. Robot, 013, 35(4): 500-51(in Chinese) (,,. [J]., 013, 35(4): 500-51) [10] Triggs B, Mclauchlan P F, Hartley R I, et al. Bundle adjustment a modern synthesis[c] //Proceedings of International Workshop on Vision Algorithms: Theory and Practice. Heidelberg: Springer, 1999: 98-37 [11] Indelman V, Williams S, Kaess M, et al. Information fusion in navigation systems via factor graph based incremental smoothing[j]. Robotics and Autonomous Systems, 013, 61(8): 71-738 [1] Forster C, Carlone L, Dellaert F, et al. IMU preintegration on manifold for efficient visual-inertial maimuma-posteriori estimation [C] //Proceedings of Robotics: Science and Systems. Rome: Robotics: Science and Systems, 015: Article No.6 [13] Leutenegger S, Lynen S, Bosse M, et al. Keyframe-based visual-inertial odometry using nonlinear optimization[j]. The International Journal of Robotics Research, 015, 34(3): 314-334 [14] Chatfield A B. Fundamentals of high accuracy inertial navigation[m]. Reston: American Institute of Astronautics and Aeronautics, 1997 [15] Lupton T, Sukkarieh S. Visual-inertial-aided navigation for high-dynamic motion in built environments without initial conditions[j]. IEEE Transactions on Robotics, 01, 8(1): 61-76 [16] Davison A J, Reid I D, Molton N D, et al. MonoSLAM: realtime single camera SLAM[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 007, 9(6):105-1067 [17] Mourikis A, Roumeliotis S. A multi-state constraint Kalman filter for vision-aided inertial navigation[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 007: 3565-357 [18] Shi J, Tomasi C. Good features to track[c] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 1994: 593-600 [19] Davison A J. Active search for real-time vision[c] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 005, 1: 66-73 [0] Moré J. The Levenberg-Marquardt algorithm: implementation and theory[j]. Numerical Analysis, 1978, 630(1): 105-116 [1] Klein G, Murray D. Parallel tracking and mapping for small AR workspaces[c] //Proceedings of IEEE and ACM International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 007: 5-34 [] Klein G, Murray D. Parallel tracking and mapping on a camera phone[c] //Proceedings of IEEE and ACM International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 009: 83-86 [3] Rosten E, Drummond T. Machine learning for high-speed corner detection[c] //Proceedings of European Conference on Computer Vision. Heidelberg: Springer, 006, 1: 430-443 [4] Huber P J. Robust statistics[m]. Hoboken: Wiley, 009 [5] Klein G, Murray D. Improving the agility of keyframe-based SLAM[C] //Proceedings of European Conference on Computer Vision. Heidelberg: Springer, 008, : 80-815 [6] Mur-Artal R, Montiel J M M, Tardos J D. ORB-SLAM: a versatile and accurate monocular SLAM system[j]. IEEE Transactions on Robotics, 015, 31(5): 1147-1163 [7] Rublee E, Rabaud V, Konolige K, et al. ORB: an efficient alternative to SIFT or SURF[C] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 011: 564-571 [8] Mur-Artal R, Tardós J D. Fast relocalisation and loop closing in keyframe-based SLAM[C] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 014: 846-853
第 6 期, 等 : 867 [9] Kümmerle R, Grisetti G, Strasdat H, et al. go: a general framework for graph optimization[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 011: 3607-3613 [30] Tan W, Liu H, Dong Z, et al. Robust monocular SLAM in dynamic environments[c] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 013: 09-18 [31] Lowe D G. Distinctive image features from scale-invariant keypoints[j]. International Journal of Computer Vision, 004, 60(): 91-110 [3] Fischler M A, Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[j]. Communications of the ACM, 1981, 4(6): 381-395 [33] Newcombe R A, Lovegrove S J, Davison A J. DTAM: dense tracking and mapping in real-time[c] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 011: 30-37 [34] Engel J, Schöps T, Cremers D. LSD-SLAM: large-scale direct monocular SLAM[C] //Proceedings of Computer Vision ECCV 014. Heidelberg: Springer, 014: 834-849 [35] Civera J, Davison A J, Montiel J M M. Inverse depth parametrization for monocular SLAM[J]. IEEE Transactions on Robotics, 008, 4(5): 93-945 [36] Chambolle A, Pock T. A first-order primal-dual algorithm for conve problems with applications to imaging[j]. Journal of Mathematical Imaging and Vision, 011, 40(1): 10-145 [37] Engel J, Sturm J, Cremers D. Semi-dense visual odometry for a monocular camera[c] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 013: 1449-1456 [38] Glover A, Maddern W, Warren M, et al. OpenFABMAP: an open source toolbo for appearance-based loop closure detection [C] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 01: 4730-4735 [39] Strasdat H, Montiel J M M, Davison A J. Real-time monocular slam: Why filter?[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 010: 657-664 [40] Sturm J, Engelhard N, Endres F, et al. A Benchmark for the evaluation of RGB-D SLAM systems[c] //Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Los Alamitos: IEEE Computer Society Press, 01: 573-580 [41] Arya S, Mount D M, Netanyahu N S, et al. An optimal algorithm for approimate nearest neighbor searching in fied dimensions[j]. Journal of the ACM, 1998, 45(6): 891-93 [4] Lovegrove S, Davison A J. Real-time spherical mosaicing using whole image alignment[c] //Proceedings of European Conference on Computer Vision. Heidelberg: Springer, 010, 3: 73-86 [43] Galvez-Lopez D, Tardos J D. Bags of binary words for fast place recognition in image sequences[j]. IEEE Transactions on Robotics, 01, 8(5): 1188-1197 [44] Schöps T, Engel J, Cremers D. Semi-dense visual odometry for AR on a smartphone[c] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 014: 145-150 [45] Forster C, Pizzoli M, Scaramuzza D. SVO: fast semi-direct monocular visual odometry[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 014: 15- [46] Nurutdinova I, Fitzgibbon A. Towards pointless structure from motion: 3D reconstruction and camera parameters from general 3D curves[c] //Proceedings of the IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 015: 363-371 [47] Tarrio J J, Pedre S. Realtime edge-based visual odometry for a monocular camera[c] //Proceedings of the IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 015: 70-710 [48] Concha A, Civera J. Using superpiels in monocular SLAM[C] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 014: 365-37 [49] Concha A, Civera J. DPPTAM: dense piecewise planar tracking and mapping from a monocular sequence[c] //Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Los Alamitos: IEEE Computer Society Press, 015: 5686-5693 [50] Salas M, Hussain W, Concha A, et al. Layout aware visual tracking and mapping[c] //Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Los Alamitos: IEEE Computer Society Press, 015: 149-156 [51] Concha A, Hussain W, Montano L, et al. Incorporating scene priors to dense monocular mapping[j]. Autonomous Robots, 015, 39(3): 79-9 [5] Furukawa Y, Ponce J. Accurate, dense, and robust multiview stereopsis[j]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 010, 3(8): 136-1376 [53] Wendel A, Maurer M, Graber G, et al. Dense reconstruction on-the-fly[c] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 01: 1450-1457 [54] Newcombe R A, Izadi S, Hilliges O, et al. KinectFusion: real-time dense surface mapping and tracking[c] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 011: 17-136 [55] Curless B, Levoy M. A volumetric method for building comple models from range images[c] //Proceedings of the 3rd Annual Conference on Computer Graphics and Inter-Active Techniques. New York: ACM Press, 1996: 303-31 [56] Lorensen W E, Cline H E. Marching cubes: a high resolution 3D surface construction algorithm[c] //Proceedings of the 14th Annual Conference on Computer Graphics and Interactive Techniques. New York: ACM Press, 1987: 163-169 [57] Yang C, Medioni G. Object modelling by registration of multiple range images[j]. Image and Vision Computing, 199, 10(3): 145-155 [58] Whelan T, Kaess M, Fallon M, et al. Kintinuous: spatially etended KinectFusion[R]. Cambridge: Massachusetts Institute of Technology, MIT-CSAIL-TR-01-00, 01 [59] Whelan T, Johannsson H, Kaess M, et al. Robust real-time visual odometry for dense RGB-D mapping[c] //Proceedings of
868 计算机辅助设计与图形学学报第 8 卷 IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 013: 574-5731 [60] Huang A S, Bachrach A, Henry P, et al. Visual odometry and mapping for autonomous flight using an RGB-D camera[c] //Proceedings of International Symposium on Robotics Research. Heidelberg: Springer, 011: Article No. [61] Steinbrücker F, Sturm J, Cremers D. Real-time visual odometry from dense RGB-D images[c] //Proceedings of IEEE International Conference on Computer Vision Workshop. Los Alamitos: IEEE Computer Society Press, 011: 719-7 [6] Whelan T, Kaess M, Leonard J J, et al. Deformation-based loop closure for large scale dense RGB-D SLAM[C] //Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Los Alamitos: IEEE Computer Society Press, 013: 548-555 [63] Sumner R W, Schmid J, Pauly M. Embedded deformation for shape manipulation[j]. ACM Transactions on Graphics, 007, 6(3): Article No.80 [64] Whelan T, Leutenegger S, Salas-Moreno R F, et al. ElasticFusion: dense SLAM without a pose graph[c] //Proceedings of Robotics: Science and Systems. Rome: Robotics: Science and Systems, 015: Article No.1 [65] Pfister H, Zwicker M, Van Baar J, et al. Surfels : surface elements as rendering primitives[c] //Proceedings of the 7th Annual Conference on Computer Graphics and Interactive Techniques. New York: ACM Press, 000: 335-34 [66] Pizzoli M, Forster C, Scaramuzza D. REMODE: probabilistic, monocular dense reconstruction in real time[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 014: 609-616 [67] Tanskanen P, Kolev K, Meier L, et al. Live metric 3D reconstruction on mobile phones[c] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 013: 65-7 [68] Kolev K, Tanskanen P, Speciale P, et al. Turning mobile phones into 3D scanners[c] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 014: 3946-3953 [69] Pradeep V, Rhemann C, Izadi S, et al. MonoFusion: real-time 3D reconstruction of small scenes with a single Web camera[c] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 013: 83-88 [70] Ondruska P, Kohli P, Izadi S. MobileFusion: real-time volumetric surface reconstruction and dense tracking on mobile phones[j]. IEEE Transactions on Visualization and Computer Graphics, 015, 1(11): 151-158 [71] Schöps T, Sattler T, Hane C, et al. 3D modeling on the go: interactive 3D reconstruction of large-scale scenes on mobile devices[c] //Proceedings of International Conference on 3D Vision. Los Alamitos: IEEE Computer Society Press, 015: 91-99 [7] Bleyer M, Rhemann C, Rother C. PatchMatch stereo-stereo matching with slanted support windows[c] //Proceedings of the British Machine Vision Conference. Guildford: BMVA Press, 011: 1-11 [73] Vogiatzis G, Hernández C. Video-based, real-time multi-view stereo[j]. Image and Vision Computing, 011, 9(7): 434-441 [74] Mur-Artal R, Tardos J D. Probabilistic semi-dense mapping from highly accurate feature-based monocular SLAM[C]// Proceedings of Robotics: Science and Systems. Rome: Robotics: Science and Systems, 015: Article No.41 [75] Li M, Mourikis A I. High-precision, consistent EKF-based visual-inertial odometry[j]. The International Journal of Robotics Research, 013, 3(6): 690-711 [76] Huang G P, Mourikis A, Roumeliotis S. Analysis and improvement of the consistency of etended Kalman filter based SLAM[C] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 008: 473-479 [77] Hesch J A, Kottas D G, Bowman S L, et al. Camera-IMU-based localization: observability analysis and consistency improvement[j]. International Journal of Robotics Research, 014, 33(1): 18-01 [78] Huang G, Kaess M, Leonard J J. Towards consistent visual-inertial navigation[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 014: 496-4933 [79] Kaess M, Johannsson H, Roberts R, et al. isam: incremental smoothing and mapping using the Bayes tree[j]. International Journal of Robotics Research, 01, 31(): 16-35 [80] Gauglitz S, Sweeney C, Ventura J, et al. Live tracking and mapping from both general and rotation-only camera motion[c] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 01: 13- [81] Pirchheim C, Schmalstieg D, Reitmayr G. Handling pure camera rotation in keyframe-based SLAM[C] //Proceedings of IEEE International Symposium on Mied and Augmented Reality. Los Alamitos: IEEE Computer Society Press, 013: 9-38 [8] Herrera C, Kim K, Kannala J, et al. DT-SLAM: deferred triangulation for robust SLAM[C] //Proceedings of International Conference on 3D Vision (3DV). Los Alamitos: IEEE Computer Society Press, 014: 609-616 [83] Hedborg J, Forssén P E, Felsberg M, et al. Rolling shutter bundle adjustment[c] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 01: 1434-1441 [84] Lovegrove S, Patron-Perez A, Sibley G. Spline fusion: a continuous-time representation for visual-inertial fusion with application to rolling shutter cameras[c] //Proceedings of the British Machine Vision Conference. Guildford: BMVA Press, 013: Article No.93 [85] Li M, Kim B H, Mourikis A I. Real-time motion tracking on a cellphone using inertial sensing and a rolling-shutter camera[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 013: 471-4719 [86] Kerl C, Stuckler J, Cremers D. Dense continuous-time tracking and mapping with rolling shutter RGB-D cameras[c] //Proceedings of IEEE International Conference on Computer Vision. Los Alamitos: IEEE Computer Society Press, 015: 64-7 [87] Kerl C, Sturm J, Cremers D. Robust odometry estimation for RGB-D cameras[c] //Proceedings of IEEE International Conference on Robotics and Automation. Los Alamitos: IEEE Computer Society Press, 013: 3748-3754