DynamicFusion: Reconstruction and tracking of non-rigid scenes in real-time
The DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Abstract
We present the first dense SLAM system capable of reconstructing non-rigidly deforming scenes in real-time, by fusing together RGBD scans captured from commodity sensors. Our DynamicFusion approach reconstructs scene geometry whilst simultaneously estimating a dense volumetric 6D motion field that warps the estimated geometry into a live frame. Like KinectFusion, our system produces increasingly denoised, detailed, and complete reconstructions as more measurements are fused, and displays the updated model in real time. Because we do not require a template or other prior scene model, the approach is applicable to a wide range of moving objects and scenes.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“In the SLAM community, Newcombe et al. [182] have address the non-rigid case for small-scale reconstruction.”
From Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016) · cites this paper as [182]
Citations
Cites
KinectFusion: Real-time dense surface mapping and tracking
Newcombe, R. A., Fitzgibbon, A., Izadi, S., Hilliges, O., Molyneaux, D., Kim, D., Davison, A. J., Kohi, P., Shotton, J., & Hodges, S. (2011) · 2011 10th IEEE International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2011.6092378 · Cited by 4,008 · cited 14×
“As in KinectFusion, volumetric TSDF memory limits geometric extent; but there are many solutions to this in the literature.”
MethodInfluentialSelf-citationA 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 3×
“The truncation distance τ > 0 and the updated TSDF value is given by the weighted averaging scheme [5], with the weight truncation introduced in [18]:”
BackgroundMethodKinectFusion: 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 13×
“As in KinectFusion, volumetric TSDF memory limits geometric extent; but there are many solutions to this in the literature.”
MethodInfluentialSelf-citationReal-time 3D reconstruction at scale using voxel hashing
Nießner, M., Zollhöfer, M., Izadi, S., & Stamminger, M. (2013) · ACM Transactions on Graphics · doi:10.1145/2508363.2508374 · Cited by 1,002 · cited 1×
“KinectFusion spurred a flurry of follow up research aimed at robustifying the tracking [9, 32] and expanding its spatial mapping capabilities to larger environments [22, 19, 34, 31, 9].”
BackgroundEfficient model-based 3D tracking of hand articulations using Kinect
Oikonomidis, I., Kyriazis, N., & Argyros, A. (2011) · Procedings of the British Machine Vision Conference 2011 · doi:10.5244/c.25.101 · Cited by 874 · cited 1×
“The best of these demonstrate high accuracy, real-time performance capture for tracking faces [16, 3], hands [21, 20], complete bodies [27], or general articulated objects [23, 33].”
BackgroundKintinuous: Spatially Extended KinectFusion
Whelan, T. J., Kaess, M., Fallon, M. F., Johannsson, H., Leonard, J. J., & McDonald, J. (2012) · Maynooth University ePrints and eTheses Archive (Maynooth University) · Cited by 624
Realtime and Robust Hand Tracking from Depth
Qian, C., Sun, X., Wei, Y., Tang, X., & Sun, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.145 · Cited by 470
Embedded deformation for shape manipulation
Sumner, R. W., Schmid, J., & Pauly, M. (2007) · ACM SIGGRAPH 2007 papers · doi:10.1145/1275808.1276478 · Cited by 449 · cited 3×
“Hierarchical Deformation Tree: In original applications of the embedded deformation graph [25] approach to non-rigid tracking, E is defined as the k−nearest neighbours of each node or all nodes within a specified radius.”
BackgroundInfluentialReal-time non-rigid reconstruction using an RGB-D camera
Zollhöfer, M., Nießner, M., Izadi, S., Rehmann, C., Zach, C., Fisher, M., Wu, C., Fitzgibbon, A., Loop, C., Theobalt, C., et al. (2014) · ACM Transactions on Graphics · doi:10.1145/2601097.2601165 · Cited by 405 · cited 1×
“Recently, [37] demonstrated an impressive real-time version of a similar technique, using GPU accelerated optimisations.”
MethodGlobal Correspondence Optimization for Non‐Rigid Registration of Depth Scans
Li, H., Sumner, R. W., & Pauly, M. (2008) · Computer Graphics Forum · doi:10.1111/j.1467-8659.2008.01282.x · Cited by 399 · cited 1×
“Practical advancements to pairwise 3D shape and scan alignment over larger deformations make use of reduced deformable model parametrisations [14, 4].”
Method
Research works printed but not linked: 1
- curless, 1997, New Methodsfor Surface Reconstructionfrom Range Images
Cited by
Items 401–500 of 956
Self-supervised Cloth Reconstruction via Action-conditioned Cloth Tracking
Huang, Z., Lin, X., & Held, D. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160653 · Cited by 22
“Numerous deformable object tracking algorithms have been developed, such as template-based tracking [26], [27], simultaneous tracking and reconstruction [28], [29], or point set registration [30], [31], [32], [33].”
MethodPhysics-Informed Neural Mapping and Motion Planning in Unknown Environments
Liu, Y., Ni, R., & Qureshi, A. H. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3548495 · Cited by 26
“These methods are grid based and are limited to a large resolution due to computational and memory costs [21], [44].”
Building High-Fidelity Human Body Models From User-Generated Data
Xu, Z., Chang, W. C., Zhu, Y., Dong, L., Zhou, H., & Zhang, Q. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3001540 · Cited by 36
“As KinectFusion is proposed under the assumption of static scenes, DynamicFusion [32] implements the real-time reconstruction of dynamic scenes under non-rigid deformation.”
MethodSpace‐Time Co‐Segmentation of Articulated Point Cloud Sequences
Yuan, Q., Li, G., Xu, K., Chen, X., & Huang, H. F. (2016) · Computer Graphics Forum · doi:10.1111/cgf.12843 · Cited by 24
“Such consistent segmentation also implies a dense correspondence across all frames of geometry, which is essential for tasks such as robust tracking [HRF13] and reconstruction [NFS15].”
MethodNRMVS: Non-Rigid Multi-View Stereo
Innmann, M., Kim, K., Gu, J., Nießner, M., Loop, C., Stamminger, M., & Kautz, J. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093583 · Cited by 17
“DynamicFusion [36] jointly optimizes a Deformation Graph [45], then fuses the deformed surface with the current depth map.”
MethodMulti-modal and inertial sensor solutions for navigation-type factor graphs
Fourie, D. (2017) · Woods Hole Open Access Server (WHOAS) · doi:10.1575/1912/9305 · Cited by 18
SimulCap : Single-View Human Performance Capture with Cloth SimulationPreprint
Yu, T., Zheng, Z., Zhong, Y., Zhao, J., Dai, Q., Pons‐Moll, G., & Liu, Y. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.06323 · Cited by 19
“tup and controlled lighting. Fusion4D [16] and Motion2Fusion [15] set up a rig with several RGBD cameras to capture dynamic scenes with challenging motions in real-time. DynamicFusion like approaches [43,25,22,55,70,56,34] reconstructed geometry and nonrigid motion simultaneously form a single-view and in realtime. Th…”
BackgroundLearning to Reconstruct People in Clothing from a Single RGB CameraPreprint
Alldieck, T., Magnor, M. A., Bhatnagar, B. L., Theobalt, C., & Pons‐Moll, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.05885 · Cited by 21
“DynamicFusion [51] generalized KinectFusion to non-rigid objects by combining non-rigid tracking and fusion.”
BackgroundDetectFusion: Detecting and Segmenting Both Known and Unknown Dynamic Objects in Real-time SLAMPreprint
Hachiuma, R., Pirchheim, C., Schmalstieg, D., & Saitô, H. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.09127 · Cited by 17
“The second approach, non-rigid object reconstruction, focuses on objects undergoing deformation, such as humans [2, 3, 13].”
Real-Time Nonrigid Mosaicking of Laparoscopy Images
Zhou, H., & Jayender, J. (2021) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3065030 · Cited by 16
“ARAP smoothing is widely used in the non-rigid SLAM systems [20], which is usually integrated into the cost function and minimized by a Gauss-Newton-like optimization method.”
BackgroundMethodInfluentialNSF: Neural Surface Fields for Human Modeling from Monocular Depth
Xue, Y., Bhatnagar, B. L., Marin, R., Sarafianos, N., Xu, Y., Pons‐Moll, G., & Tung, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01382 · Cited by 26
“Methods such as KinectFusion [48] and DynamicFusion [47] fuse depth measurements over time to create a complete and accurate model.”
MethodTotal-Recon: Deformable Scene Reconstruction for Embodied View Synthesis
Song, C., Yang, G., Deng, K., Zhu, J.-Y., & Ramanan, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01620 · Cited by 16
“Dynamic Fusion [29] builds a template-free dense SLAM sys-tem for dynamic objects.”
BackgroundChuruvija, M., Sapkota, R., Ahmed, D., & Karkee, M. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.109899 · Cited by 16
Improving 3D Reconstruction Through RGB-D Sensor Noise Modeling
Maken, F. A., Muthu, S., Nguyen, C. V., Sun, C., Tong, J., Wang, S., Tsuchida, R., Howard, D., Dunstall, S., & Petersson, L. (2025) · Sensors · doi:10.3390/s25030950 · Cited by 16
Scale-Adaptive Neural Dense Features: Learning via Hierarchical Context Aggregation
Spencer, J., Bowden, R., & Hadfield, S. (2019) · 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.2019.00636 · Cited by 20
Spatial Perception Enhancement in Assembly Training Using Augmented Volumetric Playback
Sasikumar, P., Chittajallu, S., Raj, N., Bai, H., & Billinghurst, M. (2021) · Frontiers in Virtual Reality · doi:10.3389/frvir.2021.698523 · Cited by 16
RTSDM: A Real-Time Semantic Dense Mapping System for UAVs
Li, Z., Zhao, J., Zhou, X. S., Wei, S., Li, P., & Shuang, F. (2022) · Machines · doi:10.3390/machines10040285 · Cited by 18
“Co-Fusion [37], proposed by M. Runz in 2017, used semantic information to track moving objects and reconstruct dynamic objects, like Dynamic Fusion [22].”
A Single Camera Image Based Approach for Glossy Reflections in Mixed Reality Applications
Schwandt, T., & Broll, W. (2016) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2016.12 · Cited by 16
“The approach introduced by DynamicFusion [22] might be a possible solution, which should be investigated in future work.”
BackgroundTowards Robust and Accurate Single-View Fast Human Motion Capture
Yu, T., Zhao, J., Huang, Y.-H., Li, Y., & Liu, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2920633 · Cited by 19
“[31] and [12] are the most similar work to ours in this spectrum, they combine DynamicFusion [32] like methods with a multi-person linear model, SMPL model [33], or kinematic motion constraints to reconstruct detailed surface geometry and motion of the performer.”
MethodHamiltonian Dynamics for Real-World Shape Interpolation
Eisenberger, M., & Cremers, D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58548-8_11 · Cited by 21
Multigrid Predictive Filter Flow for Unsupervised Learning on VideosPreprint
Kong, S., & Fowlkes, C. C. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.01693 · Cited by 18
“This includes tasks such as optical flow [74, 51, 102], deconvolution [43, 66, 27], non-rigid morphing [57], stereo [78, 49] defocus [45], affine alignment [42], blur removal [26], etc.”
MethodCorrespondence-Free Material Reconstruction using Sparse Surface Constraints
Weiss, S., Maier, R. L., Cremers, D., Westermann, R., & Thuerey, N. (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/cvpr42600.2020.00474 · Cited by 19
“Many non-rigid reconstruction algorithms make use of depth data in the form of a video stream to obtain geometric representations [32, 42, 52, 31, 19].”
BackgroundStrobeNet: Category-Level Multiview Reconstruction of Articulated ObjectsPreprint
Zhang, G., Litany, O., Sridhar, S., & Guibas, L. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.08016 · Cited by 16
“Tracking approaches typically operate on RGB-D videos and either assume that a template of the instance is available [23, 26, 31, 34, 43], or that neighboring frames have short baselines [44, 52, 63, 82].”
Background4D Temporally Coherent Light-Field Video
Mustafa, A., Volino, M., Guillemaut, J., & Hilton, A. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00014 · Cited by 16
LaplacianFusion: Detailed 3D Clothed-Human Body Reconstruction
Kim, H., Nam, H., Kim, J., Park, J., & Lee, S. (2022) · ACM Transactions on Graphics · doi:10.1145/3550454.3555511 · Cited by 18
“TSDF-based approaches that warp and fuse the input depth sequence onto the canonical volume have been proposed for recovering dynamic objects [Innmann et al. 2016; Newcombe et al. 2015].”
MethodEvent-Based Non-rigid Reconstruction of Low-Rank Parametrized Deformations from Contours
Xue, Y., Li, H., Leutenegger, S., & Stückler, J. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02011-z · Cited by 17
“Using RGB-D cameras simplifies the task due to the availability of dense depth for which several methods have been proposed recently [e.g. (Newcombe et al., 2015; Bozic et al., 2020)].”
Emotion information visualization through learning of 3D morphable face model
Jin, H., Wang, X., Lian, Y., & Hua, J. (2018) · The Visual Computer · doi:10.1007/s00371-018-1482-1 · Cited by 15
“[36] presented a 3D object reconstruction method which can be used for face modeling.”
MethodScene Semantic Reconstruction from Egocentric RGB-D-Thermal Videos
Luo, R., Şener, O., & Savarese, S. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00073 · Cited by 15
“More recently, DynamicFusion builds a dense SLAM system for reconstructing non-rigidly deforming scenes in real time with RGB-D information [28].”
BackgroundMethodGarmentTracking: Category-Level Garment Pose Tracking
Han, X., Xu, W., Zhang, J., Tang, T., Li, Y., Du, W., Ye, R., & Lu, C. (2023) · 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/cvpr52729.2023.02034 · Cited by 24
Safe Leaf Manipulation for Accurate Shape and Pose Estimation of Occluded Fruits
Yao, S., Pan, S., Bennewitz, M., & Hauser, K. K. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11128788 · Cited by 14
“We finally apply radial basis function interpolation [23] of our sparse deformation ˆ P ′ to a dense deformation.”
Multiple point sets registration based on Expectation Maximization algorithm
Zhou, Y., Xu, S., Jin, C., & Guo, Z. (2018) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2018.05.026 · Cited by 14
Robust point cloud registration based on both hard and soft assignments
Zhu, J., Jin, C., Jiang, Z., Xu, S., Xu, M., & Pang, S. (2018) · Optics & Laser Technology · doi:10.1016/j.optlastec.2018.07.072 · Cited by 14
Robust shape estimation for 3D deformable object manipulation
Han, T., Zhao, X., Sun, P., & Pan, J. (2018) · Communications in Information and Systems · doi:10.4310/cis.2018.v18.n2.a3 · Cited by 15
Mobile3DScanner: An Online 3D Scanner for High-quality Object Reconstruction with a Mobile Device
Xiang, X., Jiang, H., Zhang, G., Yu, Y., Li, C., Yang, X., Chen, D., & Bao, H. (2021) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2021.3106491 · Cited by 16
“DynamicFusion [23] and Fusion4D [5] employ the embedded deformation graph method [38] to track the motion of deformable objects.”
BackgroundMethodSong, S., Zhao, J., Huang, K., Lin, J., Ye, C., & Feng, T. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3396638 · Cited by 14
“Recently, research on implicit neural representations has made significant progress.”
BackgroundSceNeRFlow: Time-Consistent Reconstruction of General Dynamic Scenes
Tretschk, E., Golyanik, V., Zollhöfer, M., Božič, A., Lassner, C., & Theobalt, C. (2024) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv62453.2024.00136 · Cited by 17
“Slightly larger motions are possible with a single RGB-D camera [98], although these methods [7, 8, 10, 25, 26, 30, 43, 55, 69, 70, 97] all up-date [13] both previously unseen and previously seen parts of the canonical model over time during their online optimization, thereby losing time…”
BackgroundMethod4D Match Trees for Non-rigid Surface Alignment
Mustafa, A., Kim, H., & Hilton, A. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46448-0_13 · Cited by 14
“Partial surface tracking methods for single view [26] and RGBD data [8,27] perform sequential alignment of the reconstructions using frame-to-frame tracking.”
BackgroundMethodSurface Motion Capture Transfer with Gaussian Process Regression
Boukhayma, A., Franco, J.-S., & Boyer, E. (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.2017.379 · Cited by 14
“After more than a decade of predominance of standard motion capture, we have witnessed in the recent years the emergence of 4D surface capture systems either through high quality multi-view set-ups [9, 10] or in a more democratized form with low-cost sensors [16, 13, 27].”
BackgroundReal-time 3D scene reconstruction with dynamically moving object using a single depth camera
Lu, F., Zhou, B., Zhang, Y., & Zhao, Q. (2018) · The Visual Computer · doi:10.1007/s00371-018-1540-8 · Cited by 14
“However, for moving object, rigid transformation is not sufficient to represent the movement so that we follow DynamicFusion [16] approach and estimate the parameters of the warp field by non-rigidly deforming the canonical model to the segmented object region.”
InfluentialPreaching Voxels: An Alternative Approach to Mixed Reality
Regenbrecht, H. T., Park, J.-W. (., Ott, C., Mills, S. J., Cook, M., & Langlotz, T. (2019) · Frontiers in ICT · doi:10.3389/fict.2019.00007 · Cited by 14
“, 2017) and real-time reconstruction of dynamic scenes (Dou et al., 2013, 2016; Newcombe et al., 2015; Innmann et al., 2016).”
BackgroundAn Approach for Semantic Segmentation of Tree-like Vegetation
Digumarti, S. T., Schmid, L., Rizzi, G. M., Nieto, J. I., Siegwart, R., Beardsley, P. A., & Cadena, C. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8793576 · Cited by 17
“For example, reconstruction techniques such as [6], [7], [8] perform sub-optimally as they are forced into local minima due to small repetitive structures in the leaf-regions of a tree.”
MethodRescan: Inductive Instance Segmentation for Indoor RGBD Scans
Halber, M., Shi, Y., Xu, K., & Funkhouser, T. A. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00263 · Cited by 22
“[36] showcases a system where multiple observations of a deforming object are fused into a single consistent reconstruction.”
BackgroundVolume Sweeping: Learning Photoconsistency for Multi-View Shape Reconstruction
Leroy, V., Franco, J.-S., & Boyer, E. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-020-01377-0 · Cited by 18
Tracking Deformable Surfaces That Undergo Topological Changes Using an RGB-D Camera
Tsoli, A., & Argyros, A. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.42 · Cited by 14
“3D shape reconstruction, registration and matching: Recently a number of methods for the 3D reconstruction of dynamic geometries have been proposed [32, 3, 14, 19].”
Background4DTAM: Non-Rigid Tracking and Mapping via Dynamic Surface Gaussians
Matsuki, H., Bae, G., & Davison, A. J. (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/cvpr52734.2025.02507 · Cited by 25
“DynamicFusion [35] pioneered a line of work [13, 19, 46, 47] which captures temporal evolution in the scene geometry by jointly optimizing a canonical volumetric representation (e.g., TSDF volume [35]) and a deformation field.”
Deformable 3D Fusion: From Partial Dynamic 3D Observations to Complete 4D Models
Xu, W., Salzmann, M., Wang, Y., & Liu, Y. (2015) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.252 · Cited by 13
“, warp-field for [22] versus subspace for us.”
BackgroundTracking of Non-Rigid Objects using RGB-D Camera
Sengupta, A., Krupa, A., & Marchand, É. (2019) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2019.8914543 · Cited by 13
“[16] uses ARAP to regularize a warp field from the current state of deformation to a canonical model.”
Sarmadi, H., Muñoz‐Salinas, R., Berbis, M. A., Luna, A., & Medina-Carnicer, R. (2019) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2019.105004 · Cited by 13
Modeling Realistic Clothing From a Single Image Under Normal Guide
Liu, X., Li, J., & Lu, G. (2023) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2023.3245583 · Cited by 19
“Early KinectFusion [4] and DynamicFusion [5] are representative works.”
Background3D Data Acquisition and Registration Using Two Opposing Kinects
Soleimani, V., Mirmehdi, M., Damen, D., Hannuna, S., & Camplani, M. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.21 · Cited by 13
“Methods for capturing the full extent of an object, or a complete scene, have been proposed using handheld sensors and temporal fusion [22, 29, 49].”
BackgroundMethodInfluentialLearning Facial Expressions with 3D Mesh Convolutional Neural Network
Jin, H., Lian, Y., & Hua, J. (2018) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3200572 · Cited by 13
“[31], which can be also used for face modeling.”
BackgroundSee-your-room: indoor localization with camera vision
Sun, M., Zhang, L., Liu, Y., Miao, X., & Ding, X. X. (2019) · ACM Turing Celebration Conference - China (ACM TURC) · doi:10.1145/3321408.3321581 · Cited by 20
“Existing camera vision based approaches use vision to match pictures or features which are already known to figure out where people who are using this mobile equipment are [1, 5, 9, 10, 12, 17].”
BackgroundC urve F usion: reconstructing thin structures from RGBD sequences
Liu, L., Chen, N., Ceylan, D., Theobalt, C., Wang, W., & Mitra, N. J. (2018) · ACM Transactions on Graphics · doi:10.1145/3272127.3275097 · Cited by 13
“We would like to extend our method to also capture such dynamic structures borrowing ideas from DynamicFusion [Newcombe et al. 2015] and variations designed for a comparable surface setting.”
MethodSpatio-temporal registration of plants non-rigid 3-D structure
Zhang, T., Elnashef, B., & Filin, S. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.10.009 · Cited by 12
Korean Cattle 3D Reconstruction from Multi-View 3D-Camera System in Real Environment
Dang, C., Lee, S.-S., Alam, M., Lee, S.-M., Park, M. N., Seong, H.-S., Han, S., Nguyen, H.-P., Baek, M. K., Lee, J.-G., et al. (2024) · Sensors · doi:10.3390/s24020427 · Cited by 12
“In the case of reconstruction of non-rigid objects, using one camera to reconstruct [11,12] a moving non-rigid object has many limits due to the fast change of non-rigid objects.”
BackgroundYu, Z., Zhang, H., Liu, R., Dai, S., Chen, X., Sheng, W., & Jin, Y. (2025) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2024.101837 · Cited by 12
Frequency-Aware Uncertainty Gaussian Splatting for Dynamic Scene Reconstruction
Shao, M., Qiao, Y., Zhang, K., & Meng, L. (2025) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2025.3549143 · Cited by 12
“Following [4,35], the updated Gaussian positions μ ti can be calculated as: where R tk denotes the rotation matrix learned by the Gaussian points G k .”
VGA: Reconstructing Vivid 3D Gaussian Avatars from Monocular Videos
Liu, X., & Wu, C. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-5812-1_10 · Cited by 14
DynoSAM: Open-Source Smoothing and Mapping Framework for Dynamic SLAM
Morris, J., Wang, Y., Kliniewski, M., & Ila, V. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3641813 · Cited by 17
“However, the fundamental formulation of dynamic SLAM remains an open challenge, with no consensus on the optimal approach for accurate motion estimation within a SLAM pipeline.”
Registration of Point Clouds Based on the Ratio of Bidirectional Distances
Zhu, J., Wang, D., Bai, X., Lu, H., Jin, C., & Li, Z. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.85 · Cited by 12
“Due to the wide application in 3D reconstruction [11], shape recognition [6] and robot mapping [16, 13], point clouds registration has attracted immense attention in computer vision, pattern recognition and robotics.”
BackgroundCivera, J., & Lee, S. H. (2019) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-28603-3_6 · Cited by 12
“nertial data. 7.6 Non-rigid reconstructions The 3D reconstruction of non-rigid environments is a very relevant and challenging area of research that has been frequently addressed using RGB-D sensors. [80] is one of the most representative systems, achieving impressive results for deformable surfaces. [90] is a recent…”
BackgroundInfluentialSelf-supervised 3D Representation Learning of Dressed Humans from Social Media Videos
Jafarian, Y., & Park, H. S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3231558 · Cited by 12
“An analogous warping is used for non-rigid tracking [43] without the part based representation, which allows mapping between consecutive frames.”
MethodFH-Net: A Fast Hierarchical Network for Scene Flow Estimation on Real-World Point Clouds
Ding, L., Dong, S., Xu, T., Xu, X., Wang, J., & Li, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19842-7_13 · Cited by 20
“While there is extensive literature on traditional 3D scene flow [38,13,42,14,36,37,25], we pay more attention to recent learning-based methods[20,40,6,26,30,24] which have shown advantages in many aspects.”
BackgroundA Survey of Simultaneous Localization and Mapping with an Envision in 6G Wireless Networks
Huang, B., Zhao, J., Luo, S., & Liu, J. (2021) · Journal of Global Positioning Systems · doi:10.5081/jgps.17.2.206 · Cited by 13
FlowIBR: Leveraging Pre-Training for Efficient Neural Image-Based Rendering of Dynamic Scenes
Büsching, M., Bengtson, J., Nilsson, D., & Björkman, M. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00800 · Cited by 12
Dynamic Human Body Modeling Using a Single RGB Camera
Zhu, H., Yu, Y., Zhou, Y., & Du, S. (2016) · Sensors · doi:10.3390/s16030402 · Cited by 12
“[2] extended the KinectFusion method to capture dynamic 3D shapes that included partial views of moving people by utilizing a dense volumetric warp-field parametrization.”
MethodProactive Sensing for Improving Hand Pose Estimation
Hsiao, D.-Y., Sun, M., Ballweber, C., Cooper, S., & Popović, Z. (2016) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/2858036.2858587 · Cited by 12
Kuo, M.-Y. J., Kawahara, R., Nobuhara, S., & Nishino, K. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3075450 · Cited by 12
“space, we employ the embedded deformation graph model [20], [29], [30] to represent the non-rigid transformations for each pair of consecutive frames as a combination of a global rigid and patch-wise local affine deformations.”
BackgroundMethodMin, Z., & Meng, M. Q. ‐. (2020) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2020.3027073 · Cited by 12
TextureMe: High-Quality Textured Scene Reconstruction in Real Time
Kim, J., Kim, H., Nam, H., Park, J., & Lee, S. (2022) · ACM Transactions on Graphics · doi:10.1145/3503926 · Cited by 12
Zhao, M., Zhou, D., Song, X., Chen, X., & Zhang, L. (2022) · Sensors · doi:10.3390/s22093389 · Cited by 12
“To simultaneously achieve the accurate camera pose estimation and the dense environment reconstructions, many works have been tried, such as LSD-SLAM [12], DynamicFusion [13], and ElasticFusion [14].”
An Efficient Volumetric Mesh Representation for Real-Time Scene Reconstruction Using Spatial Hashing
Dong, W., Shi, J., Tang, W., Wang, X., & Zha, H. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8463157 · Cited by 15
DressRecon: Freeform 4D Human Reconstruction from Monocular Video
Tan, J. S., Xiang, D., Tulsiani, S., Ramanan, D., & Yang, G. (2025) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv66043.2025.00028 · Cited by 17
“Depth-based methods [10, 60, 73] follow the seminal DynamicFu-sion work [44] to integrate human shape from a monocular depth stream into a canonical space with the help of a deformation model.”
Jin, Y., Peng, S., Wang, X., Xie, T., Xu, Z., Yang, Y., Shen, Y., Bao, H., & Zhou, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01028 · Cited by 23
“Traditional methods tend to leverage complicated hardware, such as dense camera arrays [11, 17, 23, 52, 53] or depth sensors [2, 5, 37, 49, 54], to reconstruct high-fidelity human performances.”
Direct and Sparse Deformable Tracking
Lamarca, J., Rodríguez, J. J. G., Tardós, J. D., & Montiel, J. M. M. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3201253 · Cited by 11
“The first deformable SLAM method proposed was DynamicFusion [9].”
MethodLoRD: Local 4D Implicit Representation for High-Fidelity Dynamic Human Modeling
Jiang, B., Ren, X., Dou, M., Xue, X., Fu, Y., & Zhang, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19809-0_18 · Cited by 12
HMDO : Markerless multi-view hand manipulation capture with deformable objects
Xie, W., Yu, Z., Zhao, Z., Zuo, B., & Wang, Y. (2023) · Graphical Models · doi:10.1016/j.gmod.2023.101178 · Cited by 16
Transfer4D: A Framework for Frugal Motion Capture and Deformation Transfer
Maheshwari, S., Narain, R., & Hebbalaguppe, R. (2023) · 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/cvpr52729.2023.01234 · Cited by 15
“Dynamic reconstruction based methods [26, 33, 42, 43, 50] provide promising results, but tracking error could cause structural artifacts thereby resulting in issues of shape correspondence or skeleton extraction.”
BackgroundNonrigid Point Cloud Registration Using Piecewise Tricubic Polynomials as Transformation Model
Glira, P., Weidinger, C., Otepka, J., Camillo Ressl, A., Pfeifer, N., & Haberler‐Weber, M. (2023) · Remote Sensing · doi:10.3390/rs15225348 · Cited by 12
Zhu, H., Bai, S., Shi, J., Lu, J., Zuo, X., Huang, S., & Xu, Y. (2025) · The Visual Computer · doi:10.1007/s00371-025-03819-7 · Cited by 11
“DynamicFusion [23] estimates the 6D deformation field of moving objects using the warp field and reconstructs dynamic objects through deformation field fusion.”
Real-Time Enhancement of Dynamic Depth Videos with Non-Rigid Deformations
Ismaeil, K. A., Aouada, D., Solignac, T., Mirbach, B., & Ottersten, B. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2622698 · Cited by 11
“Also, similarly to KinectDeform, one may suspect DynamicFusion not to be able to performwell on a lower resolution noisier ToF camera.”
MethodSchober, M., Adam, A., Yair, O., Mazor, S., & Nowozin, 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.2017.26 · Cited by 11
“Recent work [32, 31, 30] performs 3D scene reconstruction from depth data over long temporal periods.”
BackgroundFast deformable model-based human performance capture and FVV using consumer-grade RGB-D sensors
Alexiadis, D. S., Zioulis, N., Zarpalas, D., & Daras, P. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.02.013 · Cited by 11
“With optimized GPU code, real-time operation is achieved: The recent and promising “Dynamic Fusion” (DynFu) work by [27] can reconstruct slowly deforming objects (e.”
MethodTopology-Change-Aware Volumetric Fusion for Dynamic Scene Reconstruction
Li, C., & Guo, X. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58517-4_16 · Cited by 11
3D human pose and shape estimation with dense correspondence from a single depth image
Wang, K., Zhang, G., & Yang, J. J. (2022) · The Visual Computer · doi:10.1007/s00371-021-02339-4 · Cited by 14
PREF: Predictability Regularized Neural Motion Fields
Song, L., Gong, X., Planche, B., Zheng, M., Doermann, D., Yuan, J., Chen, T., & Wu, Z. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20047-2_38 · Cited by 11
Dynamic High Resolution Deformable Articulated Tracking
Walsman, A., Wan, W., Schmidt, T., & Fox, D. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00015 · Cited by 11
Self-citationHybrid Modeling of Non-Rigid Scenes From RGBD Cameras
Malleson, C., Guillemaut, J., & Hilton, A. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2863027 · Cited by 11
“[32] note that their method cannot handle fast motion.”
BackgroundMethodShape Correspondence with Isometric and Non-Isometric Deformations
Dyke, R. M., Stride, C., Lai, Y., Rosin, P. L., Aubry, M., Boyarski, A., Bronstein, A. M., Bronstein, M. M., Cremers, D., Fisher, M., et al. (2019) · IRIS Research product catalog (Sapienza University of Rome) · doi:10.2312/3dor.20191069 · Cited by 17
“Shape correspondence is also necessary in common applications such as modelling [KMP07], reconstruction [LSP08] and tracking [NFS15].”
MethodMonoFusion: Sparse-View 4D Reconstruction via Monocular Fusion
Wang, Z., Tan, J. S., Khurana, T., Peri, N., & Ramanan, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00773 · Cited by 21
“While classical work [9, 41] often relies on RGB-D sensors, or strong domain knowledge [2, 7], recent approaches [33] based on neural radiance fields [40] have progressed towards reconstructing dynamic scenes in-the-wild from RGB video alone.”
Real-Time 3D Reconstruction of Thin Surface Based on Laser Line Scanner
He, Y., Zheng, S., Zhu, F., & Huang, X. (2020) · Sensors · doi:10.3390/s20020534 · Cited by 11
“With the help of massively parallel processors, it has become very popular in real-time 3D reconstruction [1,4,19,20,24,25].”
BackgroundParametric Model Estimation for 3D Clothed Humans from Point Clouds
Wang, K., Zheng, H., Zhang, G., & Yang, J. J. (2021) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar52148.2021.00030 · Cited by 10
“3D human shape can be reconstructed from depth sequences us-ing template-based methods [7,9,11,29,48], templateless method-s [12,25], or the combination of the two [46,47].”
BackgroundLive4D: A Real-time Capture System for Streamable Volumetric Video
Zhou, Y., Wang, S., Li, W., Zhang, C., Rao, L., Cheng, P., Xu, Y., Ke, J., Feng, W., Zhou, W., et al. (2023) · SIGGRAPH Asia Technical Communications · doi:10.1145/3610543.3626178 · Cited by 10
“We implement the embedded deformation (ED) field [Sumner et al. 2007] based on dual-quaternion blending ( DQB ), as done in [Newcombe et al. 2015].”
MethodSTDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering
Li, Z., Jiang, H., Cai, Y., Chen, J., Bi, B., Gao, S., Zhao, H., Wang, Y., Mao, T., & Wang, Z. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3836162 · Cited by 10
Real Time Stable Haptic Rendering Of 3D Deformable Streaming Surface
Tian, Y., Li, C., Guo, X., & Prabhakaran, B. (2017) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3083187.3083198 · Cited by 10
“Recent research [16] proposed the real-time method for real-time fusion of the non-rigid scene.”
Fragmentation Guided Human Shape Reconstruction
Zhang, Y., Luo, X., Yang, W., & Jingyi Yu, A. (2019) · IEEE Access · doi:10.1109/access.2019.2905879 · Cited by 10
“On the active sensing front, Newcombe et al. [4] use a single RGB-D sensor and estimate a dense volumetric 6D motion field that warps the geometry into a live frame.”
BackgroundMethodInfluentialVoxelvideos for Entertainment, Education, and Training
Regenbrecht, H. T., Ott, C., Park, N., Duncan, S. P., & Collins, J. (2021) · IEEE Access · doi:10.1109/access.2021.3076488 · Cited by 10
“Voxel data structures are often used internally; Microsoft’s KinectFusion [16], [17] uses voxels for internal data representation and RemixedReality [18] use a voxel grid to store spatial and temporal transformations which are applied to the virtual environment, where the virtual environment is…”
MethodHuman Motion Tracking with Less Constraint of Initial Posture from a Single RGB-D Sensor
Liu, C., Wang, A., Bu, C., Wang, W., & Sun, H. (2021) · Sensors · doi:10.3390/s21093029 · Cited by 11
“Please refer to [3,46] for Sensors 2021 , 21 , x FOR PEER REVIEW 9 of 16 the scheme more detail.”
BackgroundMethodInfluentialLarge-Scale Dense Mapping System Based on Visual-Inertial Odometry and Densely Connected U-Net
Fan, C., Hou, J., & Yu, L. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3250301 · Cited by 10
“Numerous RGB-D SLAM methods, such as KinectFusion [1], [2], DynamicFusion [3], Kintinuous [4], [5], [6], and Elasticfusion [7], use depth cameras for 6DoF pose estimation.”
BackgroundMethodSemanticFusion: Joint Labeling, Tracking and Mapping
Cavallari, T., & Stefano, L. D. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-49409-8_55 · Cited by 10
“Recent works, instead, have demonstrated impressive results in non-rigid and dynamic settings by modeling motion as a fundamental property of the captured scene [3,14].”
BackgroundResult
Metadata
- Authors
- Newcombe, R. A., Fox, D., & Seitz, S. M.
- Affiliations
- University of Washington (US)
- Venue
- 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2015-06
- Pages
- 343-352
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1938204631
- Cited by
- 951
- References
- 42
- Citations by year
- 2026: 28 · 2025: 61 · 2024: 67 · 2023: 93 · 2022: 84 · 2021: 113 · 2020: 105 · 2019: 124 · 2018: 110 · 2017: 78 · 2016: 80 · 2015: 8
- FWCI
- 159.57
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Keywords
- dense SLAM, real-time 3D reconstruction, KinectFusion
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.