KinectFusion: Real-time dense surface mapping and tracking
The DOI matches its Crossref record; 32 of the 34 references it cites were found in a registry or an index.
Abstract
We present a system for accurate real-time mapping of complex and arbitrary indoor scenes in variable lighting conditions, using only a moving low-cost depth camera and commodity graphics hardware. We fuse all of the depth data streamed from a Kinect sensor into a single global implicit surface model of the observed scene in real-time. The current sensor pose is simultaneously obtained by tracking the live depth frame relative to the global model using a coarse-to-fine iterative closest point (ICP) algorithm, which uses all of the observed depth data available. We demonstrate the advantages of tracking against the growing full surface model compared with frame-to-frame tracking, obtaining tracking and mapping results in constant time within room sized scenes with limited drift and high accuracy. We also show both qualitative and quantitative results relating to various aspects of our tracking and mapping system. Modelling of natural scenes, in real-time with only commodity sensor and GPU hardware, promises an exciting step forward in augmented reality (AR), in particular, it allows dense surfaces to be reconstructed in real-time, with a level of detail and robustness beyond any solution yet presented using passive computer vision.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
A method for registration of 3-D shapes
Besl, P. J., & McKay, N. D. (1992) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.121791 · Cited by 18,103 · cited 2×
“The most important class of algorithms have been based on the ICP concept introduced in [3] which poses data alignment as a nonlinear optimisation problem in which correspondences between scans are approximated using the closest pairs of points found between scans at the previous iteration.”
BackgroundMethodMarching cubes: A high resolution 3D surface construction algorithm
Lorensen, W. E., & Cline, H. E. (1987) · Proceedings of the 14th annual conference on Computer graphics and interactive techniques · doi:10.1145/37401.37422 · Cited by 10,104 · cited 1×
“One option is to extract the connected surfaces using a marching cubes type algorithm [18], followed by a standard rasterising rendering pipeline.”
MethodBilateral Filtering for Gray and Color Images
Tomasi, C., & Manduchi, R. (2002) · Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271) · doi:10.1109/iccv.1998.710815 · Cited by 8,177 · cited 1×
“We apply a bilateral filter [27] to the raw depth map to obtain a discontinuity preserved depth map with reduced noise Dk,”
MethodParallel Tracking and Mapping for Small AR Workspaces
Klein, G., & Murray, D. (2007) · 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2007.4538852 · Cited by 4,346 · cited 3×
“By using only depth data, our system can work in complete darkness mitigating any issues concerning low light conditions, problematic for passive camera [17, 19, 26] and RGB-D based systems [14].”
BackgroundMethodObject modelling by registration of multiple range images
Chen, Y., & Medioni, G. (1992) · Image and Vision Computing · doi:10.1016/0262-8856(92)90066-c · Cited by 3,154
A volumetric method for building complex models from range images
Curless, B., & Levoy, M. (1996) · Proceedings of the 23rd annual conference on Computer graphics and interactive techniques · doi:10.1145/237170.237269 · Cited by 3,124 · cited 6×
“A related non-parametric representation used in graphics is the signed distance function (SDF) introduced in [7] for the purpose of fusing partial depth scans while mitigating problems related to mesh-based reconstruction algorithms.”
BackgroundMethodInfluentialA Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms
Seitz, S. M., Curless, B., Diebel, J., Scharstein, D., & Szeliski, R. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06) · doi:10.1109/cvpr.2006.19 · Cited by 2,457 · cited 1×
“In computer vision, research on structure from motion (SFM) and multi-view stereo (MVS) has produced many compelling re-sults, in particular accurate camera tracking and sparse reconstructions (e.g. [10]), and increasingly reconstruction of dense surfaces (e.g. [24]).”
BackgroundIterative point matching for registration of free-form curves and surfaces
Zhang, Z. (1994) · International Journal of Computer Vision · doi:10.1007/bf01427149 · Cited by 2,402 · cited 1×
“A number of ICP variants perform early iterations on a subset of possibly corresponding points or operate within a coarse-to-fine scheme [31], speeding up both the data association and final pose optimisation.”
BackgroundKinectFusion: real-time 3D reconstruction and interaction using a moving depth camera
Izadi, S., Kim, D., Hilliges, O., Molyneaux, D., Newcombe, R., Kohli, P., Shotton, J., Hodges, S., Freeman, D., Davison, A., et al. (2011) · Proceedings of the 24th annual ACM symposium on User interface software and technology · doi:10.1145/2047196.2047270 · Cited by 2,110 · cited 1×
“In [16] we discuss all these possibilities in detail.”
BackgroundSelf-citationDTAM: Dense tracking and mapping in real-time
Newcombe, R. A., Lovegrove, S. J., & Davison, A. J. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126513 · Cited by 2,078
Self-citation
Research works printed but not linked: 4
- weise, 0, Inhand scanning with online loop closure, Proceedings of the IEEE International Workshop on 3D Digital Imaging and Modeling (3DIM) 2009
- wurm, 0, OctoMap: A probabilistic, flexible, and compact 3D map representation for robotic systems, Proceedings of the ICRA Workshop on Best Practice in 3D Perception and Modeling for Mobile Manipulation 2010
- 0, NVIDIA OptiX 2 Ray Tracing Engine
- zach, 0, A globally optimal algorithm for robust TV-L1 range image integration, Proceedings of the International Conference on Computer Vision (ICCV) 2007
Cited by
Items 701–800 of 1,006
Semi-direct EKF-based monocular visual-inertial odometry
Tanskanen, P., Nägeli, T., Pollefeys, M., & Hilliges, O. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7354242 · Cited by 61
“1) Dense methods: Dense direct methods operate on surface measurements directly, either depth estimates of a stereo camera or a RGB-D sensor [8], [15], and do not extract sets of features from this data.”
BackgroundMethodSelf-citationEvaluation of Multi-view 3D Reconstruction Software
Schöning, J., & Heidemann, G. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-23117-4_39 · Cited by 61
“For 3D reconstruction, various approaches such as 123D Catch [3], PhotoScan [2], Photo tourism [23], VideoTrace [11], Kinect fusion [17], ProFORMA [18] etc.”
MethodContour-based next-best view planning from point cloud segmentation of unknown objects
Monica, R., & Aleotti, J. (2017) · Autonomous Robots · doi:10.1007/s10514-017-9618-0 · Cited by 61
Semantic Multi-view Stereo: Jointly Estimating Objects and Voxels
Ulusoy, A. O., Black, M. J., & Geiger, A. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.482 · Cited by 61
“Volumetric Reconstruction from Images: While there is a large body of literature on volumetric fusion from range images [10,25], in this paper we focus on reconstruction directly from RGB images.”
BackgroundOasis: Procedurally Generated Social Virtual Spaces from 3D Scanned Real Spaces
Sra, M., Garrido-Jurado, S., & Maes, P. (2017) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2017.2762691 · Cited by 61
“KinectFusion [9] and its variants [10], [11] are among the most popular techniques for”
MethodPolarimetric Dense Monocular SLAM
Yang, L., Tan, F., Li, A., Cui, Z., Furukawa, Y., & Tan, P. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00406 · Cited by 61
“For robustness and accuracy, fusion algorithms use RGBD cameras [27, 6, 40], which however increase power consumptions and limit the operating capabilities to short-range indoor scenes.”
Robust 3D Self-Portraits in Seconds
Li, Z., Yu, T., Pan, C., Zheng, Z., & Liu, Y. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00142 · Cited by 61
OmniSLAM: Omnidirectional Localization and Dense Mapping for Wide-baseline Multi-camera Systems
Won, C., Seok, H., Cui, Z., Pollefeys, M., & Lim, J. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196695 · Cited by 61
“To this end, many range sensors, for example, laser-based LiDAR [1] and structured light 3D scanners [2] can be used as their depth sensing is accurate, which is critical to the mapping results.”
BackgroundRecurrent Multi-view Alignment Network for Unsupervised Surface Registration
Feng, W., Zhang, J., Cai, H., Xu, H., Hou, J., & Bao, H. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.01016 · Cited by 61
“It has wide applications in many fields such as 3D reconstruction [32, 31], tracking [33, 48] and medical imaging [63, 39].”
Overview of Multi-Robot Collaborative SLAM from the Perspective of Data Fusion
Chen, W., Wang, X., Gao, S., Shang, G., Zhou, C., Li, Z., Xu, C., & Hu, K. (2023) · Machines · doi:10.3390/machines11060653 · Cited by 61
“Recent studies have shown that many niche sensors can also be used for SLAM, such as olfactory [41], thermal [42], magnetic [43], event camera [50], luminous depth camera [51], and light field camera [52] sensors.”
BackgroundImMesh: An Immediate LiDAR Localization and Meshing Framework
Lin, J., Yuan, C., Cai, Y., Li, H., Ren, Y., Zou, Y., Hong, X., & Zhang, F. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2023.3321227 · Cited by 60
HI-SLAM2: Geometry-Aware Gaussian SLAM for Fast Monocular Scene Reconstruction
Zhang, W., Cheng, Q., Skuddis, D., Zeller, N., Cremers, D., & Haala, N. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3626627 · Cited by 61
“LiDAR systems require expensive hardware setups and an additional camera for capturing color information, while RGB-D sensors suffer from limited operational range and sensitive to varying lighting conditions.”
Exploring High-Level Plane Primitives for Indoor 3D Reconstruction with a Hand-held RGB-D Camera
Dou, M., Guan, L., Frahm, J., & Fuchs, H. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-37484-5_9 · Cited by 60
“Recent approaches [1][2][3][4] demonstrated the use of a hand-held RGB-D camera for indoor 3D reconstructions.”
MethodResultInfluentialRise of the Indoor Crowd: Reconstruction of Building Interior View via Mobile Crowdsourcing
Chen, S., Li, M., Ren, K., Fu, X., & Qiao, C. (2015) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/2809695.2809702 · Cited by 60
“Other works like [13, 26] uses Kinect, a commercial depth sensor to reconstruct 3D model.”
MethodMulti-Channel Generalized-ICP: A robust framework for multi-channel scan registration
Servos, J., & Waslander, S. L. (2016) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2016.10.016 · Cited by 59
“Newcombe et al. also proposed the dense SLAM method KinectFusion [17], which uses a Kinect sensor to generate dense 3D maps of small areas.”
MethodInfluentialMontage4D: interactive seamless fusion of multiview video textures
Du, R., Chuang, M., Chang, W., Hoppe, H., & Varshney, A. (2018) · ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D) · doi:10.1145/3190834.3190843 · Cited by 60
“KinectFusion [Izadi et al. 2011; Newcombe et al. 2011] is the first system that tracks and fuses point clouds into dense meshes using a single depth sensor.”
BackgroundCavallari, T., Bertinetto, L., Mukhoti, J., Torr, P. H. S., & Golodetz, S. (2019) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2019.00068 · Cited by 60
“…this formulation does not take into account the practical requirements on a camera relocaliser for live scenarios such as interactive dense SLAM [53], in which it is infeasible to spend hours or even days training a relo-caliser on the scene of interest; rather, a relocaliser must be trained…”
BackgroundAlignment of the Virtual Scene to the Tracking Space of a Mixed Reality Head-Mounted DisplayPreprint
Azimi, E., Qian, L., Navab, N., & Kazanzides, P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.05834 · Cited by 59
“Marker-based tracking algorithms have the advantage in simplicity and robustness [9, 19], and markerfree tracking algorithms offer better user experience [18, 22].”
BackgroundContrastive 3D Shape Completion and Reconstruction for Agricultural Robots Using RGB-D Frames
Magistri, F., Marks, E., Nagulavancha, S., Vizzo, I., Läbe, T., Behley, J., Halstead, M., McCool, C., & Stachniss, C. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3193239 · Cited by 58
“We start by registering the RGB-D images to each other with a standard TSDF fusion pipeline [12], [37].”
MethodH$_{2}$-Mapping: Real-Time Dense Mapping Using Hierarchical Hybrid Representation
Jiang, C., Zhang, H., Liu, P., Yu, Z., Cheng, H., Zhou, B., & Shen, S. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3313051 · Cited by 58
“Previous works utilize explicit scene representations like occupancy grids [1], TSDF [2], [3], [4], [5], [6], [7], surfels [8], [9], and meshes [10] to achieve real-time performance.”
MethodSparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering
Sun, C., Choe, J., Loop, C., Ma, W.-C., & Wang, Y.-C. F. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01509 · Cited by 58
Hu, L., Ma, C., Luo, L., Wei, L., & Li, H. (2014) · ACM Transactions on Graphics · doi:10.1145/2661229.2661254 · Cited by 58
“The point cloud can be derived from the vertices of a mesh reconstructed by a consumer depth camera [Newcombe et al. 2011] or directly from multi-view stereo [Furukawa and Ponce 2010].”
MethodJoint Detection, Tracking and Mapping by Semantic Bundle Adjustment
Fioraio, N., & Stefano, L. D. (2013) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2013.202 · Cited by 58
“The system builds on established tracking and mapping techniques to exploit incremental 3D reconstruction in order to validate hypotheses on the presence and pose of sought objects.”
MethodAn Integrative Framework of Human Hand Gesture Segmentation for Human–Robot Interaction
Ju, Z., Ji, X., Li, J., & Liu, H. (2015) · IEEE Systems Journal · doi:10.1109/jsyst.2015.2468231 · Cited by 58
“Its broad applications cover 3-D reconstruction [4], [5], image processing [6], human–machine interface [2], [7]–[10], robotics”
BackgroundInstant Mixed Reality Lighting from Casual Scanning
Richter-Trummer, T., Kalkofen, D., Park, J., & Schmalstieg, D. (2016) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2016.18 · Cited by 58
“Kinect Fusion [18], a popular SLAM variant, turns RGB-D input into a volumetric model, from which a surface mesh is extracted using isosurface raycasting.”
MethodInfluentialScene flow propagation for semantic mapping and object discovery in dynamic street scenes
Kochanov, D., Ošep, A., Stückler, J., & Leibe, B. (2016) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2016.7759285 · Cited by 58
“In the indoor RGB-D SLAM domain, KinectFusion [10] and pointbased fusion techniques [11] have been proposed that can separate dynamic parts from the static background and exclude them from tracking and mapping.”
MethodDense Visual SLAM with Probabilistic Surfel Map
Yan, Z., Ye, M., & Ren, L. (2017) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2017.2734458 · Cited by 58
Effects of volumetric capture avatars on social presence in immersive virtual environments
Cho, S., Kim, S., Lee, J., Ahn, J., & Han, J. (2020) · IEEE Conference on Virtual Reality and 3D User Interfaces (VR) · doi:10.1109/vr46266.2020.00020 · Cited by 58
Extrinsic Calibration from Per-Sensor Egomotion
Brookshire, J., & Teller, S. J. (2012) · Robotics: Science and Systems · doi:10.15607/rss.2012.viii.004 · Cited by 58
“We projected the noise into h and, using a chi-squared goodness-of-fit test, found the projected Gaussian to be a good approximation (at 5% significance) for both FOVIS and KinectFusion.”
MethodInfluentialOVIR-3D: Open-Vocabulary 3D Instance Retrieval Without Training on 3D DataPreprint
Lu, S., Chang, H., Jing, E., Boularias, A., & Bekris, K. E. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2311.02873 · Cited by 58
“Early works have addressed the 3D reconstruction problem either through online SLAM meth-ods [22, 23, 24, 25, 26] or offline methods like structure-from-motion [27, 28] using a variety of 3D representations, such as TSDF [29], Surfel [30], and more recently NeRF [31, 32, 33, 34].”
BackgroundMethodSajad Saeedi, G., Spink, T., Gorgovan, C., Webb, A. M., Clarkson, J., Tomusk, E., Debrunner, T., Kaszyk, K., Marugán, P. G. D. A., Rodchenko, A., et al. (2018) · Proceedings of the IEEE · doi:10.1109/jproc.2018.2856739 · Cited by 57
Self-citationImproving Depth Completion via Depth Feature Upsampling
Wang, Y., Zhang, G., Wang, S., Li, B., Liu, Q., Hui, L., & Dai, Y. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01994 · Cited by 56
“Accurate and dense scene depth is crucial for various applications [13, 34, 41], such as autonomous navigation [33] and augmented reality [24].”
BackgroundActiveGS: Active Scene Reconstruction Using Gaussian Splatting
Jin, L., Zhong, X., Pan, Y., Behley, J., Stachniss, C., & Popović, M. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3555149 · Cited by 56
“For the mesh evaluation, we run TSDF fusion [23] on depth images rendered at training viewpoints and extract the scene mesh using Marching Cubes [20].”
A Memory-Efficient KinectFusion Using Octree
Zeng, M., Zhao, F., Zheng, J., & Liu, X. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-34263-9_30 · Cited by 56
“Especially, Newcombe et al. [9] and Izadi et al. [6] proposed KinectFusion, which used a commodity depth camera Kinect[8] to scan and model the dense surface of a room-size (about 4 m × 4 m × 4 m ) scene in realtime.”
BackgroundMethodInfluentialEfficient onbard RGBD-SLAM for autonomous MAVs
Scherer, S., & Zell, A. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696482 · Cited by 56
“There are also some approaches that rely on depth images alone, e.g. KinectFusion [3], which uses coarse-to-fine iterative closest point with projective data association implemented on graphics hardware for registration with the map.”
MethodWize Mirror - a smart, multisensory cardio-metabolic risk monitoring system
Andreu, Y., Chiarugi, F., Colantonio, S., Giannakakis, G. A., Giorgi, D., Henríquez, P., Kazantzaki, E., Manousos, D., Kostas Marias, A., Matuszewski, B. J., et al. (2016) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2016.03.018 · Cited by 56
Kin2. A Kinect 2 toolbox for MATLAB
Terven, J. R., & Esparza, D. M. C. (2016) · Science of Computer Programming · doi:10.1016/j.scico.2016.05.009 · Cited by 56
Wang, Y., Zhang, S., Wan, B., He, W., & Bai, X. (2018) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-018-2575-8 · Cited by 56
“At the same time, the tracking accuracy of our method is contrasted with other point cloud-based methods (KinectFusion [22] and RGBD-ICP The results of the experiment are shown in Fig.”
MethodLeaf Area Estimation by Semantic Segmentation of Point Cloud of Tomato Plants
Masuda, T. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00159 · Cited by 56
“For each voxel V , we determine the signed distance field (SDF) representation [15, 6, 18], which is a set of [ ¯ X V , N V , ¯ C V ] , where ¯ X V and ¯ C V are the averages of X V = { X | X 2 V } and C V = { C | C 2 V } , respectively.”
MethodFit2Form: 3D Generative Model for Robot Gripper Form DesignPreprint
Ha, H., Agrawal, S., & Song, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.06498 · Cited by 56
“In contrast, our model is able to efficiently explore the full 3D geometric space by representing 3D geometry with Truncated Signed Distance Functions (TSDF) [14].”
A Template-Based 3D Reconstruction of Colon Structures and Textures From Stereo Colonoscopic Images
Zhang, S., Zhao, L., Huang, S., Ye, M., & Hao, Q. (2020) · IEEE Transactions on Medical Robotics and Bionics · doi:10.1109/tmrb.2020.3044108 · Cited by 55
“Currently, the popularization of visual SLAM algorithms which can be classified into sparse [15], [16], semi-dense [17], [18], [19], [20] and dense reconstruction [21], [22], [23] have inspired researchers to apply SLAM technology to recover the 3D structures of the human colon.”
BackgroundMethodSampling-Based Path Planning for High-Quality Aerial 3D Reconstruction of Urban Scenes
Yan, F., Xia, E., Li, Z., & Zhou, Z. (2021) · Remote Sensing · doi:10.3390/rs13050989 · Cited by 55
“Combined with an RGB-D based SLAM system [26], Xu et al.”
MethodOn 3D Reconstruction Using RGB-D Cameras
Tychola, K. A., Tsimperidis, I., & Papakostas, G. A. (2022) · Digital · doi:10.3390/digital2030022 · Cited by 56
“The literature search was achieved by identifying the appropriate criteria used by other papers published in peer-reviewed journals or peer-reviewed conference proceedings, with keywords according to the title in English.”
MethodBNV-Fusion: Dense 3D Reconstruction using Bi-level Neural Volume Fusion
Li, K., Tang, Y., Prisacariu, V. A., & Torr, P. H. S. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00607 · Cited by 55
“Microsoft Kinect [45], Apple LiDAR scanner [1]) has enabled researchers to develop reconstruction algorithms taking advantage of depth maps [9, 26, 33].”
InfluentialWang, G., Pan, L., Peng, S., Liu, S., Xu, C., Miao, Y., Zhan, W., Tomizuka, M., Pollefeys, M., & Wang, H. (2025) · The International Journal of Robotics Research · doi:10.1177/02783649251374246 · Cited by 59
“TSDF-based methods recover surfaces by extracting the zero-level set, naturally capturing scene geometry with high sharpness and accuracy (Newcombe et al. 2011; Bylow et al. 2013).”
Efficient compositional approaches for real-time robust direct visual odometry from RGB-D data
Klose, S., Heise, P., & Knoll, A. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696487 · Cited by 55
“The KinectFusion [17], [18] algorithm uses a similar approach for the scene representation, but is designed to work with RGBD sensors.”
MethodInfluentialRobust RGB-D simultaneous localization and mapping using planar point features
Gao, X., & Zhang, T. (2015) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2015.03.007 · Cited by 56
Combining Inertial Navigation and ICP for Real-time 3D Surface Reconstruction
Nießner, M., Dai, A., & Fisher, M. (2014) · Eurographics · doi:10.2312/egsh.20141003 · Cited by 55
Data-driven contextual modeling for 3D scene understanding
Shi, Y., Long, P., Xu, K., Huang, H. F., & Xiong, Y. (2015) · Computers & Graphics · doi:10.1016/j.cag.2015.11.003 · Cited by 55
Physically consistent state estimation and system identification for contacts
Kolev, S., & Todorov, E. V. (2015) · IEEE-RAS International Conference on Humanoid Robots (Humanoids) · doi:10.1109/humanoids.2015.7363481 · Cited by 55
“problem is very similar to ours [7] [8].”
MethodResultBodin, B., Nardi, L., Zia, M. Z., Wagstaff, H., Shenoy, G. S., Emani, M. K., Mawer, J., Kotselidis, C., Nisbet, A. P., Luján, M., et al. (2016) · International Conference on Parallel Architectures and Compilation Techniques (PACT) · doi:10.1145/2967938.2967963 · Cited by 55
“One such dense SLAM algorithm is KinectFusion [33] (KFusion) which estimates the pose of a depth camera whilst constructing a highly detailed 3D model of the environment.”
BackgroundMethodInfluentialRGBDTAM: A cost-effective and accurate RGB-D tracking and mapping system
Concha, A., & Civera, J. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206593 · Cited by 55
“We follow the standard approach of Parallel Tracking and Mapping, first proposed in [21], and divide our algorithm into two threads.”
MethodHashimoto, T., & Saito, M. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · Cited by 55
Online Continuous Mapping using Gaussian Process Implicit Surfaces
Lee, B., Zhang, C., Huang, Z., & Lee, D. D. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794324 · Cited by 55
“Besides OG, SDF have been employed in many studies ([10], [11], [12]).”
MethodSelf-Supervised Deep Depth Denoising
Sterzentsenko, V., Saroglou, L., Chatzitofis, A., Thermos, S., Zioulis, N., Doumanoglou, A., Zarpalas, D., & Daras, P. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00133 · Cited by 55
Deformation-Aware 3D Model Embedding and Retrieval
Uy, M. A., Huang, J., Sung, M., Birdal, T., & Guibas, L. J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58571-6_24 · Cited by 55
Robust Visual-Lidar Simultaneous Localization and Mapping System for UAV
Qian, J., Chen, K., Chen, Q., Yang, Y., Zhang, J., & Chen, S. (2021) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2021.3099166 · Cited by 55
PanoDiffusion: 360-degree Panorama Outpainting via Diffusion
Wu, T., Zheng, C., & Cham, T. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2307.03177 · Cited by 56
“An obvious limitation, however, is that capturing, collecting and restoring such extensive datasets with 360° images is a high-effort and high-cost undertaking [1, 2], while manually creating a 3D space from scratch can be a demanding task [18, 4, 24].”
NaVILA: Legged Robot Vision-Language-Action Model for Navigation
Cheng, A.-C., Ji, Y., Yang, Z., Gongye, Z., Zou, X., Kautz, J., Biyik, E., Yin, H., Liu, S., & Wang, X. (2025) · Robotics: Science and Systems XXI · doi:10.15607/rss.2025.xxi.018 · Cited by 54
Topologically-Aware Deformation Fields for Single-View 3D Reconstruction
Duggal, S., & Pathak, D. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00159 · Cited by 56
“Reconstructing 3D from 2D observations has been an actively studied problem [9,30,43,45,53,57].”
BackgroundMethodLearning-Based Multi-View Stereo: A Survey
Wang, F., Zhu, Q., Chang, D., Gao, Q., Han, J., Zhang, T., Hartley, R., & Pollefeys, M. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3654665 · Cited by 58
“3, a typical depth map-based MVS pipeline mainly consists of feature extraction (Section III-A), cost volume construction (Section III-B), cost volume regularization (Section III-C), and depth estimation (Section III-E).”
Pose Based Person Identification Using Kinect
Sinha, A., & Chakravarty, K. (2013) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2013.91 · Cited by 54
“Not only that, none of them have given any insight how they took care of inherent noise [17] in the Kinect raw skeleton data.”
BackgroundIntegrated perception, mapping, and footstep planning for humanoid navigation among 3D obstacles
Maier, D., Lutz, C., & Bennewitz, M. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696731 · Cited by 54
“[17] presented Kinect Fusion, an approach for dense surface modeling and pose tracking with RGB-D cameras.”
MethodRobust Tracking for Real-Time Dense RGB-D Mapping with Kintinuous
Whelan, T. J., Johannsson, H., Kaess, M., Leonard, J. J., & McDonald, J. (2012) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 54
Joint 3D Object and Layout Inference from a Single RGB-D Image
Geiger, A., & Wang, C. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24947-6_15 · Cited by 54
“While the problem has traditionally been approached using RGB images [1, 8, 23, 36, 46] and videos [42], the availability of RGB-D sensors [30] and datasets [38] nourish the hope for more accurate models of the scene [11, 16, 18, 47].”
BackgroundFast planar surface 3D SLAM using LIDAR
Lenac, K., Kitanov, A., Cupec, R., & Petrovic, I. K. (2017) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2017.03.013 · Cited by 54
Surface Reconstruction Assessment in Photogrammetric Applications
Nocerino, E., Stathopoulou, E. K., Rigon, S., & Fabio Remondino, A. (2020) · Sensors · doi:10.3390/s20205863 · Cited by 54
“• Method 3: Given the image poses, a mesh model is generated from the depth maps produced in Method 1, employing a volume integration approach [15,18,19].”
InfluentialTexMesh: Reconstructing Detailed Human Texture and Geometry from RGB-D Video
Zhi, T., Lassner, C., Tung, T., Stoll, C., Narasimhan, S. G., & Vo, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58607-2_29 · Cited by 54
NIS-SLAM: Neural Implicit Semantic RGB-D SLAM for 3D Consistent Scene Understanding
Zhai, H., Huang, G., Hu, Q., Li, G., Bao, H., & Zhang, G. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3456201 · Cited by 53
“Taking advantage of RGB-D cameras, some traditional approaches use iterative closest points (ICP) [1] and truncated signed distance function (TSDF) [11] to achieve good 3D reconstruction results, such as KinectFusion [42], BundleFusion [13], and ElasticFusion [69].”
MethodPmomo: Projection Mapping on Movable 3D Object
Zhou, Y., Xiao, S., Tang, N., Wei, Z., & Xu, C. G. (2016) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/2858036.2858329 · Cited by 53
Efficient Global Point Cloud Alignment Using Bayesian Nonparametric Mixtures
Straub, J., Campbell, T., How, J. P., & Fisher, J. S. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.258 · Cited by 53
“As such, popular local optimization techniques suffice only in circumstances with small true relative transformations and large overlap, such as in dense 3D incremental mapping [23, 40, 54].”
BackgroundAutonomous reconstruction of unknown indoor scenes guided by time-varying tensor fields
Xu, K., Zheng, L., Yan, Z., Yan, G., Zhang, E., Nießner, M., Deußen, O., Cohen‐Or, D., & Huang, H. F. (2017) · ACM Transactions on Graphics · doi:10.1145/3130800.3130812 · Cited by 53
“, [Chen et al. 2013; Dai et al. 2016; Newcombe et al. 2011; Nießner et al. 2013; Whelan et al. 2015]), autonomous scene scanning and dense 3D reconstruction of indoor environments by mobile robots has drawn increasing attention from robotics and graphics communities [Charrow et al.”
BackgroundAn Overview of Humanoid Robots Technologies
Stasse, O., & Flayols, T. (2018) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-93870-7_13 · Cited by 53
“When it is possible to embed a graphical processing unit (GPU) it is possible to obtain locally dense maps in real time [47].”
BackgroundA Survey of Simultaneous Localization and Mapping.
Huang, B., Zhao, J., & Liu, J. (2019) · Cited by 53
SCFusion: Real-time Incremental Scene Reconstruction with Semantic Completion
Wu, S.-C., Tateno, K., Navab, N., & Tombari, F. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00090 · Cited by 53
“For TSDF-based methods, Kinect Fusion [34] relies on a 3D volume which stores a Truncated Signed Distance Field (TSDF) value at each voxel.”
BackgroundMethodInfluentialRobust Depth Completion with Uncertainty-Driven Loss Functions
Zhu, Y., Dong, W., Li, L., Wu, J., Li, X., & Shi, G. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i3.20275 · Cited by 55
Marks, E., Magistri, F., & Stachniss, C. (2022) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra46639.2022.9811358 · Cited by 52
“…via the Federal Office for Agriculture and Food (BLE) under the innovation support programme under funding no 28DK108B20 (RegisTer). additionally prevents the use of most volumetric environment representations commonly used in SLAM systems, such as voxel grids or TSDFs [13], [22], [27], [28], [31].”
MethodA Review of Recurrent Neural Network Based Camera Localization for Indoor Environments
Alam, M. S., Mohamed, F., Selamat, A., & Hossain, A. B. (2023) · IEEE Access · doi:10.1109/access.2023.3272479 · Cited by 52
“KinectFusion: The first RGB-D sensor-based algorithm to function in real-time was the KinectFusion algorithm In maximum medium-sized spaces, the Kinect-Fusion algorithm can map robots effectively in Robotics [195].”
MethodPanoptic Mapping with Fruit Completion and Pose Estimation for Horticultural Robots
Pan, Y., Magistri, F., Läbe, T., Marks, E., Smitt, C., McCool, C., Behley, J., & Stachniss, C. (2023) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros55552.2023.10342067 · Cited by 52
“However, in the agricultural context, classical volumetric mapping pipelines, such as KinectFusion [21] or Voxblox [22], yield incomplete fruits representations as they do not deal explicitly with the presented occlusions.”
BackgroundVistadream: Sampling Multiview Consistent Images for Single-View Scene Reconstruction
Wang, H., Liu, Y., Liu, Z., Wang, W., Dong, Z., & Yang, B. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02485 · Cited by 52
“Traditionally, this has required multiple images from different viewpoints [49] or specialized hardware like RGBD scanners [39, 61] to capture both geometry and appearance.”
Analysis and Evaluation Between the First and the Second Generation of RGB-D Sensors
Gesto, M., Tombari, F., Rodríguez‐Gonzálvez, P., & González‐Aguilera, D. (2015) · IEEE Sensors Journal · doi:10.1109/jsen.2015.2459139 · Cited by 52
“Indeed, these devices have notably pushed forward the research activity in 3D computer vision and robotic perception, and they have now replaced traditional sensors for applications such as 3D object recognition [1], Simultaneous Localization and Mapping (SLAM) [2] and 3D reconstruction [3]–[5].”
BackgroundAutonomous Exploration of Urban Environments using Unmanned Aerial Vehicles
Adler, B., Xiao, J., & Zhang, J. (2014) · Journal of Field Robotics · doi:10.1002/rob.21526 · Cited by 52
“Just two years ago, algorithms were presented, capable of localizing robots with six degrees of freedom, while mapping three-or even four-dimensional space in real-time (Newcombe et al., 2011).”
BackgroundInfluentialDepth camera tracking with contour cues
Zhou, Q.-Y., & Koltun, V. (2015) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2015.7298662 · Cited by 52
“The influential KinectFusion system [16] demonstrated real-time depth camera tracking and dense scene reconstruction by registering incoming depth images to a volumetric representation of the scene.”
MethodTowards precise real-time 3D difference detection for industrial applications
Kahn, S., Bockholt, U., Kuijper, A., & Fellner, D. W. (2013) · Computers in Industry · doi:10.1016/j.compind.2013.04.004 · Cited by 52
“Section 4 gives an overview of approaches for reducing depth measurement inaccuracies and describes the integration of a 3D reconstruction algorithm [10][11] into the 3D difference detection pipeline.”
BackgroundMethodInfluentialLive tracking and mapping from both general and rotation-only camera motion
Gauglitz, S., Sweeney, C., Ventura, J., Turk, M., & Höllerer, T. (2012) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2012.6402532 · Cited by 52
“In particular, active depth sensors based on time-of-flight or structured light [19, 23] have recently generated significant interest and can arguably provide for more detailed models and more robust tracking when lighting conditions for vision-based tracking are unfavorable.”
BackgroundPractical and accurate calibration of RGB-D cameras using spheres
Staranowicz, A. N., Brown, G. R., Morbidi, F., & Mariottini, G.-L. (2015) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2015.03.013 · Cited by 52
STAR3D: Simultaneous Tracking and Reconstruction of 3D Objects Using RGB-D Data
Ren, C. Y., Prisacariu, V. A., Murray, D. W., & Reid, I. D. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.197 · Cited by 52
“Quantitative comparison between camera pose output when using KinectFusion[7] and our method.”
BackgroundMethodResultInfluentialA Versatile Learning-Based 3D Temporal Tracker: Scalable, Robust, Online
Tan, D. J., Tombari, F., Ilić, S., & Navab, N. (2015) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.86 · Cited by 52
“Most works [2, 11, 12, 15] have applied ICP as an integral component of their algorithms; while, others [9, 16, 19, 22] have developed it to different extensions.”
MethodMulti-scale assembly with robot teams
Doğar, M. R., Knepper, R. A., Spielberg, A., Choi, C., Christensen, H. I., & Rus, D. L. (2015) · The International Journal of Robotics Research · doi:10.1177/0278364915586606 · Cited by 52
“Petersen et al. (2011) present a system where mobile robots can build structures and climb the structures they build.”
BackgroundAn Efficient Global Point Cloud Descriptor for Object Recognition and Pose Estimation
Lima, J. P., & Teichrieb, V. (2016) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi.2016.017 · Cited by 52
“As future work, pose refinement speed may be improved with a GPU implementation of ICP, such as in [22].”
BackgroundModel-Driven Feedforward Prediction for Manipulation of Deformable Objects
Li, Y., Wang, Y., Yue, Y., Xu, D., Case, M., Chang, S.-F., Grinspun, E., & Allen, P. K. (2018) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2017.2766228 · Cited by 44
SparseFusion: Dynamic Human Avatar Modeling From Sparse RGBD Images
Zuo, X., Wang, S., Zheng, J., Yu, W., Gong, M., Yang, R., & Cheng, L. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3001506 · Cited by 52
“As an extension to the celebrated KinectFusion [36] system, a dynamic fusion [35] approach has been developed which takes non-rigid motion into account by solving a non-rigid warp field for every frame.”
MethodSPSG: Self-Supervised Photometric Scene Generation from RGB-D Scans
Dai, A., Siddiqui, Y., Thies, J., Valentin, J., & Nießner, M. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00179 · Cited by 52
“RGB-D 3D Reconstruction 3D reconstruction of objects and scenes using RGB-D data is a well explored field [21, 15, 31, 6].”
BackgroundHuang, J., Zhang, Y., & Sun, M. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01506 · Cited by 52
“State-of-the-art algorithms [47, 28, 65, 12] or commercial softwares [1] ease the reconstruction and digitization of the real-world environments.”
BackgroundRuhkamp, P., Gao, D., Chen, H., Navab, N., & Busam, B. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00092 · Cited by 52
Lan-grasp: Using Large Language Models for Semantic Object Grasping and PlacementPreprint
Mirjalili, R., Krawez, M., Blei, Y., Silenzi, S., Walter, F., & Burgard, W. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.05239 · Cited by 52
Ground Truth Data, Content, Metrics, and Analysis
Krig, S. (2014) · Apress eBooks · doi:10.1007/978-1-4302-5930-5_7 · Cited by 52
ImaginateAR: AI-Assisted In-Situ Authoring in Augmented Reality
Lee, J., Aleotti, F., Mazala, D., Garcia-Hernando, G., Vicente, S., Johnston, O., Kraus-Liang, I., Powierza, J., Shin, D.-H., Froehlich, J. E., et al. (2025) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3746059.3747635 · Cited by 52
“Typically, this is achieved in two steps: first, a 3D map of the environment is built using cameras [67, 91], sometimes augmented with depth or IMU sensors [25, 68].”
Castellani, U., & Bartoli, A. (2012) · Springer eBooks · doi:10.1007/978-1-4471-4063-4_6 · Cited by 51
“The reverse calibration approach is especially effective for real-time modelling purposes [125, 30, 102].”
BackgroundMethodInfluentialRobust 3D Reconstruction With an RGB-D Camera
Wang, K., Zhang, G., & Bao, H. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2352851 · Cited by 51
“erated results are distorted due to large registration error using both Kinectfusion [5] and loop closure constraint between the first and last frames.”
BackgroundMethodInfluential
Metadata
- Authors
- Newcombe, R. A., Fitzgibbon, A., Izadi, S., Hilliges, O., Molyneaux, D., Kim, D., Davison, A. J., Kohi, P., Shotton, J., & Hodges, S.
- Affiliations
- Imperial College London (GB)
- Microsoft (United States) (US)
- Venue
- 2011 10th IEEE International Symposium on Mixed and Augmented Reality
- Publisher
- IEEE
- Published
- 2011-10
- Pages
- 127-136
- ISSN
- 1554-7868 (ISSN-L)
- OpenAlex
- W1987648924
- Cited by
- 4,008
- References
- 34
- Citations by year
- 2026: 120 · 2025: 200 · 2024: 213 · 2023: 252 · 2022: 231 · 2021: 299 · 2020: 314 · 2019: 357 · 2018: 383 · 2017: 361 · 2016: 322 · 2015: 343 · 2014: 256 · 2013: 243 · 2012: 112
- FWCI
- 378.08
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR (publishedVersion)
- Lancaster EPrints (Lancaster University) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › 3D Shape Modeling and Analysis
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Keywords
- KinectFusion, real-time 3D reconstruction, depth camera, Kinect sensor, iterative closest point, pose tracking, indoor scene reconstruction, augmented reality, GPU rendering
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.