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 601–700 of 956
Review of Research on Multi-robot Visual Simultaneous Localization and Mapping
Hesheng, Y., Shuo, P., Xu, L., & Bo, H. (2022) · Journal of Mechanical Engineering · doi:10.3901/jme.2022.11.011 · Cited by 8
Volumetric Avatar Reconstruction with Spatio-Temporally Offset RGBD Cameras
Rendle, G., Kreskowski, A., & Froehlich, B. (2023) · IEEE Conference on Virtual Reality and 3D User Interfaces (VR) · doi:10.1109/vr55154.2023.00023 · Cited by 6
“Real-time template-free methods achieve smooth reconstruction quality by leveraging temporal volumetric fusion to blend surface geometry observed in previous frames with that of the current frame [10, 11, 20, 24, 36, 38].”
BackgroundMethodInfluentialMono-STAR: Mono-Camera Scene-Level Tracking and Reconstruction
Chang, H., Ramesh, D. M., Geng, S., Gan, Y., & Boularias, A. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160778 · Cited by 7
“[6] Volume Deform SurfelWarp [10] TCAFusion [11] Co-fusion [3] Fusion4D [12] Motion2Fusion [13] Functon4D [14] MaskFusion [4] RigidFusion [15] MidFusion [5] OcclusionFusion [7] STAR-no-prior [2] Mono-STAR [16], [17] is the problem of reconstructing the geometry and recording the deformation of a…”
BackgroundMethodSAILOR: Synergizing Radiance and Occupancy Fields for Live Human Performance Capture
Dong, Z., Xu, K., Gao, Y., Sun, Q., Bao, H., Xu, W., & Lau, R. W. H. (2023) · ACM Transactions on Graphics · doi:10.1145/3618370 · Cited by 6
“The deformation fields typically map the observed coordinates to canonical coordinates of a template of the target, following the non-rigid reconstruction-and-tracking scheme [Newcombe et al. 2015b].”
BackgroundMethodAccurate 3D face modeling and recognition from RGB-D stream in the presence of large pose changes
Kim, D., Choi, J., Leksut, J. T., & Medioni, G. G. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532912 · Cited by 6
Latent-Smoothness Nonrigid Structure From Motion by Revisiting Multilinear Factorization
Dong, Q., & Wang, H. (2018) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2849146 · Cited by 6
Dynamic 3D reconstruction improvement via intensity video guided 4D fusion
Zhang, J., Maniatis, C., Horna, L., & Fisher, R. B. (2018) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2018.07.007 · Cited by 6
“In contrast to 4D fusion based on intensity motion fields for 3D/depth noise reduction, there is a group of algorithms that directly generate volumetric 6D motion fields {Ri,Ti} using depth data from Kinect sensors and reconstruct improved 3D scenes via dense 3D/depth frame registration, such as KinectFusion [13], Dyn…”
BackgroundMethodKunert, C., Schwandt, T., & Broll, W. (2018) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2018.00023 · Cited by 6
“Other approaches extend KinectFusion to reconstruct dynamic scenes [22].”
BackgroundDepth perception of moving objects viaing structured light sensor with unstructured grid
Wang, H., Li, D., Wu, C., & Yu, X. (2019) · Results in Physics · doi:10.1016/j.rinp.2019.102163 · Cited by 6
3D Reconstruction from RGB-D Data
Malleson, C., Guillemaut, J., & Hilton, A. (2019) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-28603-3_5 · Cited by 7
CNN-Based Denoising, Completion, and Prediction of Whole-Body Human-Depth Images
Jang, J. W., Kwon, Y. C., Lim, H., & Choi, O. (2019) · IEEE Access · doi:10.1109/access.2019.2957862 · Cited by 6
“Because the pose change and deformation of the models are performed in low-dimensional parametric space instead of voxel space [7], model-to-frame tracking becomes more robust.”
BackgroundMethodInfluentialKernel learning for visual perception
Wang, C. (2019) · doi:10.32657/10220/47835 · Cited by 11
RT3DHVC: A Real-Time Human Holographic Video Conferencing System With a Consumer RGB-D Camera Array
Jiang, Y., Li, J., Dai, Y., Qin, H., Zhou, X., Zhang, Y., Liu, H., Yan, K., & Yang, T. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3520621 · Cited by 5
“Cameras based on structured light, which actively project infrared light spots, can obtain high-precision depth information [1], [22].”
NRST: Non-rigid Surface Tracking from Monocular Video
Habermann, M., Xu, W., Rhodin, H., Zollhöfer, M., Pons‐Moll, G., & Theobalt, C. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-12939-2_23 · Cited by 5
Spatio-Temporal Human Shape Completion With Implicit Function Networks
Zhou, B., Franco, J.-S., Bogo, F., & Boyer, E. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00076 · Cited by 5
“Dynamic-Fusion [30] reconstructs 3D information by using the fusion of multiview depth image, which is not in our case of single view completion.”
Liu, X., Li, J., & Lu, G. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3381937 · Cited by 5
“Multi-View-based Clothing Reconstruction: Usually leverages RGB-D data [24], [25], [26] to calculate the signed distance fields of the subject, and then extract the human body and clothing mesh based on the Marching Cube [27] method.”
Gaussian Surfel Splatting for Live Human Performance Capture
Dong, Z., Xu, K., Gao, Y., Bao, H., Xu, W., & Lau, R. W. H. (2024) · ACM Transactions on Graphics · doi:10.1145/3687993 · Cited by 5
“The motions in dynamic scenes are typically modeled by learning the deformation fields [Park et al. 2021a; Peng et al. 2023; Pumarola et al. 2021; Tretschk et al. 2021] in a non-rigid reconstruction-and-tracking manner [Newcombe et al. 2015b].”
Volumetric Mapping with Panoptic Refinement using Kernel Density Estimation for Mobile Robots
Nguyen, K., Dang, T., & Huber, M. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802224 · Cited by 6
Spatio-Temporal Consistent Mapping of Growing Plants for Agricultural Robots in the Wild
Lobefaro, L., Malladi, M. V. R., Guadagnino, T., & Stachniss, C. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802321 · Cited by 5
“Although several non-rigid registration methods have been proposed to work with noisy RGB-D sensors [15], [20], [42], none explore the same setting in agriculture.”
Prediction Maps for Real-Time 3D Footstep Planning in Dynamic Environments
Karkowski, P., & Bennewitz, M. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8793999 · Cited by 7
“Finally, a variety of methods exist that compute the piecewise motion for RGB-D images or 3D LiDAR data to estimate the scence flow [14], [15], or semantically classify the image in static and dynamic parts [16].”
MethodDRACO: Weakly Supervised Dense Reconstruction And Canonicalization of Objects
Sajnani, R., Sanchawala, A., Murthy, J. K., Sridhar, S., & Krishna, K. M. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561100 · Cited by 5
“Examples include dense simultaneous localization and mapping [19, 20], manipulation [21], and non-rigid tracking [22].”
BackgroundProbability re-weighted 3D point cloud registration for missing correspondences
Sun, Z., Zhang, R., Hu, J., & Liu, X. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12134-5 · Cited by 6
“It plays an indispensable role in numerous vision systems, such as 3D mapping [18, 21], simultaneous localization and mapping (SLAM) [19], dense scene reconstruction [28], etc.”
Background4D Temporally Coherent Multi-Person Semantic Reconstruction and Segmentation
Mustafa, A., Russell, C., & Hilton, A. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01599-4 · Cited by 6
“Previous work (Newcombe et al. 2015; Mustafa et al. 2017) showed that sparse key-frames allow robust long-term correspondence for 4D reconstruction.”
BackgroundImproving RGB-D-based 3D reconstruction by combining voxels and points
Liu, X., Li, J., & Lu, G. (2022) · The Visual Computer · doi:10.1007/s00371-022-02661-5 · Cited by 5
CrossHuman: Learning Cross-guidance from Multi-frame Images for Human Reconstruction
Chen, L., Li, J., Huang, H., & Guo, Y. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548351 · Cited by 5
MP‐NeRF: Neural Radiance Fields for Dynamic Multi‐person synthesis from Sparse Views
Chao, X., & Leung, H. W. H. (2022) · Computer Graphics Forum · doi:10.1111/cgf.14646 · Cited by 6
Wrinkles Realistic Clothing Reconstruction by Combining Implicit and Explicit Method
Liu, X., Li, J., & Lu, G. (2023) · Computer-Aided Design · doi:10.1016/j.cad.2023.103514 · Cited by 5
Shi, Y., Tao, Y., Yang, M., Liu, Y., Yi, L., & Zhou, B. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3400830 · Cited by 6
“Early works like dynamic fusion [18] and rigid fusion [19] used RGB-D sensors to capture depth maps for geometry reconstruction in dynamic objects.”
Tan, F., Fu, C., Deng, T., Cai, J., & Cham, T. (2017) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3123266.3123281 · Cited by 5
“Very recently, the Dynam-icFusion [23] and VolumeDeform [15] systems demonstrated the superiority of volumetric representation for real-time dynamic reconstruction.”
BackgroundGlobal as-Conformal-as-Possible Non-Rigid Registration of Multi-view Scans
Wu, Z., Li, K., Lai, Y., & Yang, J. (2019) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2019.00061 · Cited by 5
“[7] present a system that reconstructs non-rigidly deforming subjects in real-time.”
BackgroundFurukawa, R., Nagamatsu, G., & Kawasaki, H. (2019) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2019.00057 · Cited by 5
“To achieve this, simultaneous registration and shape fusion of sequentially scanned 3D data have been extensively researched and developed [26, 21, 22].”
BackgroundSingle RGB-D Fitting: Total Human Modeling with an RGB-D Shot
Fang, X., Yang, J., Rao, J., Wang, L., & Deng, Z. (2019) · ACM Symposium on Virtual Reality Software and Technology (VRST) · doi:10.1145/3359996.3364252 · Cited by 6
Real-Time Surface Deformation Recovery from Stereo Videos
Zhou, H., & Jayender, J. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32239-7_38 · Cited by 5
“e parameters of the control points to make the deformed template match the output of the GPU-based stereo matching method. The algorithms are parallelized and run on the GPU. Similar to DynamicFusion [4], we employ dual-quaternion to represent deformation and each control point i is assigned a dual-quaternion Wt i to…”
MethodZhang, Y., Zhao, L., & Huang, S. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9197410 · Cited by 5
“DynamicFusion [14] extends KinectFusion [15] and proposes a 3D reconstruction framework for deformable object observed by an RGB-D sensor.”
MethodDense Correspondences between Human Bodies via Learning Transformation Synchronization on Graphs
Huang, X., Yang, H., Vouga, E., & Huang, Q. (2020) · neural information processing systems · Cited by 8
Autonomous Tracking For Volumetric Video Sequences
Moynihan, M., Ruano, S., Pagés, R., & Smolić, A. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00170 · Cited by 5
“Some approaches acknowledge this problem and instead opt for the use of an evolving, canonical model which is constructed over the course of the capture [6, 23, 35].”
BackgroundTactile computing: Essential building blocks for the Tactile Internet
Aßmann, U., Baier, C., Dubslaff, C., Grzelak, D., Hanisch, S., Hartono, A. P. P., Köpsell, S., Lin, T., & Strufe, T. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-821343-8.00025-3 · Cited by 11
Scene-level Tracking and Reconstruction without Object Priors
Chang, H., & Boularias, A. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9982159 · Cited by 5
“For this step, we principally follow the method of DynamicFusion [13] and SurfelWarp [12], and adapt them to our surfelbased model and multi-view setting.”
BackgroundMethodInfluentialA Visual Programming Interface for Experimenting with Volumetric Video
Hogue, A., Poremba, C., Szkudlarek, V., Fox-Gieg, N., Quevedo, A. U., Orian, C., & Anderson, J. E. (2022) · IEEE Games, Entertainment, Media Conference (GEM) · doi:10.1109/gem56474.2022.10017777 · Cited by 5
“Various non-rigid tracking algorithms such as DynamicFusion [4], [14], SobelevFusion [22], and PoseFusion [10] exist to reduce data size and track non-rigid deformations, however, These algorithms, tend to be closed-source, or unavailable.”
BackgroundA navigation paradigm driven classification for video-based rendering techniques
dos Anjos, R. K., Pereira, J. M., & Gaspar, J. A. (2018) · Computers & Graphics · doi:10.1016/j.cag.2018.10.017 · Cited by 5
“In controlled environments this can be 702 done by using markers detected by the camera [80] [9], but on 703 dynamic environments the most common approach is to track 704 features using structure from motion [11, 81, 82], providing a 705”
MethodNon-iterative RGB-D-inertial OdometryPreprint
Wang, C., Hoang, M.-C., Xie, L., & Yuan, J. (2017) · arXiv (Cornell University) · Cited by 5
TightCap: 3D Human Shape Capture with Clothing TightnessPreprint
Chen, X., Pang, A., Yang, W. H., Xui, L., & Jingyi Yu, A. (2019) · arXiv (Cornell University) · Cited by 5
Deep Learning: A Primer for Neurosurgeons
Hong-xi, Y., Yuwen, C., Cheng, X., Fan, H., Wang, X., & Ge, Z. (2024) · Advances in experimental medicine and biology · doi:10.1007/978-3-031-64892-2_4 · Cited by 5
EGG-Fusion: Efficient 3D Reconstruction with Geometry-aware Gaussian Surfel on the Fly
Pan, X., Li, Z., Ye, Z., Zhai, H., & Zhang, G. (2025) · SIGGRAPH Asia Conference Papers · doi:10.1145/3757377.3763878 · Cited by 5
Learning Pose Controllable Human Reconstruction With Dynamic Implicit Fields From a Single Image
Li, J., Liu, X., & Lu, G. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3363493 · Cited by 4
MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo
Fei, T., Bi, L., Gao, J., Chen, S., & Zhang, G. (2024) · Complex & Intelligent Systems · doi:10.1007/s40747-024-01691-x · Cited by 4
“Visual Representation: In terms of 3D field representation, the most frequent approaches are based on point [39]; mesh [40] and voxel [41] as well as parametric surfaces [42]; implicit surfaces [43] and neural radiance field drawing [10].”
Multiview Point Cloud Registration via Optimization in an Autoencoder Latent Space
Vedrenne, L., Faisan, S., & Fortun, D. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3565998 · Cited by 4
“D OWNSTREAM tasks in 3D computer vision often involve a rigid registration step [1], [2], [3], [4], [5], [6], which consists of determining 6D rigid transformations to align objects.”
GaPT-DAR: Category-level Garments Pose Tracking via Integrated 2D Deformation and 3D Reconstruction
Zhang, L., Xu, M., Wang, J., Yu, Q., Yang, L., Li, Y., Lu, C., Wang, R., & Liu, L. (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/cvpr52734.2025.02108 · Cited by 8
Topology-Aware Optimization of Gaussian Primitives for Human-Centric Volumetric Videos
Jiang, Y., Guo, C., Wu, Y., Hong, Y., Zhu, S., Shen, Z., Zhang, Y., Jiao, S., Su, Z., Xu, L., et al. (2025) · SIGGRAPH Asia Conference Papers · doi:10.1145/3757377.3763975 · Cited by 4
“To handle complex topological changes in multi-view performance capture, we employ a dynamic Gaussian representation that enables both efficient compression and high-quality rendering.”
Model-Based Real-Time Non-Rigid Tracking
Bronte, S., Bergasa, L. M., Pizarro, D., & Barea, R. (2017) · Sensors · doi:10.3390/s17102342 · Cited by 4
“On the other hand, Kinect fusion was further extended to handle deformable objects in dynamic fusion [47] for depth cameras.”
MethodControllable Variation Synthesis for Surface Motion Capture
Boukhayma, A., & Boyer, E. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00043 · Cited by 4
“4D surface capture is an emerging tool for acquiring 3D dynamic shape models of moving objects either with high quality multi-view set-ups [10, 11, 32, 31, 33] or with affordable low-cost sensors [37, 26, 17].”
MethodEstimating Human Shape Under Clothing from Single Frontal View Point Cloud of a Dressed Human
Wang, J., Lu, Z., & Liao, Q. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803008 · Cited by 4
“Methods [16, 17, 18] are model-free methods and based on free-form deformation.”
MethodZhuravlev, A. V., Razevig, V. V., & Ge, D. (2019) · IEEE International Conference on Microwaves, Communications, Antennas and Electronic Systems (COMCAS) · doi:10.1109/comcas44984.2019.8958282 · Cited by 4
“The examples of relevant computer vision methods can be found in [7], [8].”
3D Facial Emotion Recognition Using Deep Learning Technique
Rao, P., Choudhary, A., & Kumar, V. (2019) · Review of Computer Engineering Studies · doi:10.18280/rces.060303 · Cited by 4
“[3] which can also be used in facial mapping, introduced recent advances in RGBD cameras to make the 3D test faster, less costly and more precise.”
BackgroundA New Volumetric Fusion Strategy with Adaptive Weight Field for RGB-D Reconstruction
Liu, X., Li, J., & Lu, G. (2020) · Sensors · doi:10.3390/s20154330 · Cited by 4
“Subsequently, co bining this fusion method with a deformation graph [14], different methods suc as DynamicFusion [15], Fusion4D [16], KillingFusion [7], and SobolevFusion [17] realize the real-time reconstruction of the dynamic scene, which greatly expands the application scope of this kind of method.”
MethodZampokas, G., Peleka, G., Tsiolis, K., Topalidou‐Kyniazopoulou, A., Mariolis, I. G., & Tzovaras, D. K. (2022) · Medical Physics · doi:10.1002/mp.15830 · Cited by 4
RGBD-based Real-time Volumetric Reconstruction System: Architecture Design and Implementation
Zhou, K., Guo, S., Hu, J., Wang, J., Wang, Q., & Song, L. (2022) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip56404.2022.10008839 · Cited by 4
Generating High-Fidelity Texture in RGB-D Reconstruction using Patches Density Regularization
Liu, X., Li, J., & Lu, G. (2023) · Computer-Aided Design · doi:10.1016/j.cad.2023.103516 · Cited by 4
Kernel-GPA: A globally optimal solution to deformable SLAM in closed-form
Bai, F., Wu, K., & Bartoli, A. (2023) · The International Journal of Robotics Research · doi:10.1177/02783649231195380 · Cited by 4
“…representations, e.g., the Free-Form Deformations (FFD) Rueckert, Sonoda, Hayes, Hill, Leach and Hawkes (1999); Szeliski and Coughlan (1997), the Radial Basis Functions (RBF) Bookstein (1989); Fornefett, Rohr and Stiehl (2001) and the Thin-Plate Spline (TPS) Duchon (1976); Bookstein (1989).”
BackgroundCherdchusakulchai, R., Thoumrungroje, S., Tungpanjasil, T., Pimpin, A., Srituravanich, W., & Damrongplasit, N. (2024) · IEEE Access · doi:10.1109/access.2024.3414658 · Cited by 4
“By iteratively refining the 3D model to minimize this energy function, a detailed and accurate human model can be obtained [10], [11].”
MethodDynoSurf: Neural Deformation-Based Temporally Consistent Dynamic Surface Reconstruction
Yao, Y., Ren, S., Hou, J., Deng, Z., Zhang, J., & Wang, W. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73414-4_16 · Cited by 4
Filter Flow Made Practical: Massively Parallel and Lock-Free
Ravi, S. N., Xiong, Y., Mukherjee, L., & Singh, V. (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.2017.532 · Cited by 4
“The representative problems in this class correspond to a number of core topics in modern vision literature: optical flow[30, 24, 38], deconvolution [21, 27], non-rigid morphing [25], stereo (plus its variations)[34, 23, 37], defocus [22] and so on — these are all distinct problems but at the high level, deal with est…”
BackgroundCapture of hair geometry using white structured light
Chen, Y., Song, Z. P., Lin, S., Martin, R. R., & Cheng, Z. (2017) · Computer-Aided Design · doi:10.1016/j.cad.2017.10.006 · Cited by 4
Non-rigid surface recovery with a robust local-rigidity prior
Cha, G., Lee, M., Cho, J., & Oh, S. (2018) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.03.019 · Cited by 4
Skeleton Driven Non-Rigid Motion Tracking and 3D Reconstruction
Elanattil, S., Moghadam, P., Denman, S., Sridharan, S., & Fookes, C. (2018) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2018.8615797 · Cited by 4
Improved Iterative Closest Point Algorithm using Truncated Signed Distance Function
Kim, H., Hong, H., & Lee, B. H. (2018) · International Conference on Control, Automation and Systems (ICCAS) · Cited by 4
A Literature Review: Geometric Methods and Their Applications in Human-Related Analysis
Gong, W., Zhang, B., Wang, C., Yue, H., Li, C., Xing, L., Qiao, Y., Zhang, W., & Gong, F. (2019) · Sensors · doi:10.3390/s19122809 · Cited by 4
Robust Template-Based Non-Rigid Motion Tracking Using Local Coordinate Regularization
Li, W., Zhao, S., Xiao, X., & Hahn, J. K. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093533 · Cited by 5
“DynamicFusion [22] uses dual quaternions to represent deformations for better quality of transformation blending.”
BackgroundMethodSelf-supervised Depth Denoising Using Lower- and Higher-quality RGB-D sensors
Shabanov, A., Krotov, I., Chinaev, N., Poletaev, V., Kozlukov, S., Pasechnik, I., Yakupov, B., Sanakoyeu, A., Lebedev, V., & Ulyanov, D. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00084 · Cited by 4
“Low signal-to-noise ratio of mobile depth sensors prevents the use of the existing well-developed computer vision methods [27, 26, 6, 47] that are designed to work with more accurate sensors, such as Microsoft Kinect V2 [33].”
MethodMoCo‐Flow: Neural Motion Consensus Flow for Dynamic Humans in Stationary Monocular Cameras
Chen, X., Li, W., Cohen‐Or, D., Mitra, N. J., & Chen, B. (2022) · Computer Graphics Forum · doi:10.1111/cgf.14465 · Cited by 4
Wang, X., Prévost, S., Boukhayma, A., Desjardin, É., Loscos, C., Morisset, B., & Multon, F. (2023) · Computers & Graphics · doi:10.1016/j.cag.2023.07.005 · Cited by 4
Xu, W., Du, W., Han, X., Li, Y., Ye, R., Wang, Y., & Lu, C. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00012 · Cited by 7
“A milestone work is DynamicFu-sion [34], which is the first approach to achieve non-rigid reconstruction in real-time.”
InfluentialZhang, B., Li, J., Deng, X., Zhang, Y., Ma, C., & Wang, H. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01312 · Cited by 9
“It plays a key role in many applications such as shape reconstruction [29, 35, 25], texture mapping [33, 10], and shape editing [10, 9, 39].”
BackgroundEfficient Model-Free Anthropometry from Depth Data
Probst, T., Fossati, A., Salzmann, M., & Gool, L. V. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00062 · Cited by 4
“There is a vast literature on multi-view 3D and 4D shape reconstruction from multiple cameras [14, 16, 17, 46], multiple depth sensors or depth sequences [33, 5, 21, 20, 44, 2, 45].”
BackgroundAn Interactive 4D Vision Augmentation of Rapid Motion
Tao, T., Ratsamee, P., Uranishi, Y., Kiyokawa, K., Mashita, T., & Takemura, H. (2018) · Augmented Human International Conference · doi:10.1145/3174910.3174934 · Cited by 4
Human Shape Capture and Tracking at Home
Mishra, G. K., Saini, S., Varanasi, K., & Narayanan, P. J. (2018) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2018.00049 · Cited by 4
“Newcombe et al. [28] is able to reason about the canonical shape topology while reconstructing dense motion.”
BackgroundStructured Skip List: A Compact Data Structure for 3D Reconstruction
Li, S., Cheng, M., Liu, Y., Lu, S.-P., Wang, Y., & Prisacariu, V. A. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8594075 · Cited by 4
“DynamicFusion [15] aims to handle the reconstruction of non-rigid scenes.”
BackgroundA Novel Joint Points and Silhouette-Based Method to Estimate 3D Human Pose and Shape
Li, Z., Heyden, A., & Oskarsson, M. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-68763-2_4 · Cited by 4
PoseFusion2: Simultaneous Background Reconstruction and Human Shape Recovery in Real-time
Zhang, H., Zhang, T., Lam, T. L., & Vijayakumar, S. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636658 · Cited by 4
“Overall, dense key-point tracking based non-rigid dynamic human reconstructing methods such as [19] [12] [13] are able to show more surface details e.”
Self-supervised CNN for Unconstrained 3D Facial Performance Capture from an RGB-D Camera.Preprint
Guo, Y., Zhang, J., Cai, L., Cai, J., & Zheng, J. (2018) · arXiv (Cornell University) · Cited by 4
Live Reconstruction of Large-Scale Dynamic Outdoor WorldsPreprint
Mikšík, O., & Vineet, V. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.06708 · Cited by 5
“This has been addressed by similar systems [27, 12], however, such ap-”
MethodTowards 3D Human Shape Recovery Under Clothing.Preprint
Chen, X., Pang, A., Zhu, Y., Li, Y., Luo, X., Zhang, G., Wang, P., Zhang, Y., Li, S., & Jingyi Yu, A. (2019) · arXiv (Cornell University) · Cited by 4
Unsupervised Metric Relocalization Using Transform Consistency LossPreprint
Kasper, M., Nobre, F. D., Heckman, C., & Keivan, N. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.00608 · Cited by 4
“for learning robust feature descriptors [12]. Training data is sourced from multiple, independent 3D reconstructions of a single subject, built using a 3D reconstruction pipeline (i.e. DynamicFusion [26]). They then train solely on intra-sequence correspondences. Despite not training on inter-sequence correspondences,…”
MethodA Unified Optimization Framework for Low-Rank Inducing Penalties
Örnhag, M. V., & Olsson, C. (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/cvpr42600.2020.00850 · Cited by 3
“In the formulation of [12], where rank 1 corresponds to a rigid scene this can be thought of as an “as rigid as possible” prior, which is realistic in many cases [31, 24, 33, 18], but which has yet to be considered in the context of factorization methods.”
BackgroundNeRFDeformer: NeRF Transformation from a Single View via 3D Scene Flows
Tang, Z., Ren, Z., Zhao, X., Wen, B., Tremblay, J., Birchfield, S. T., & Schwing, A. G. (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/cvpr52733.2024.00980 · Cited by 3
“Online non-rigid tracking methods [3, 11, 27] explore linear blending of anchor points as a flow design, but they do not apply their flow on NeRF-based new view synthesis.”
BackgroundDynamic View Synthesis From Small Camera Motion Videos
Sun, H., Li, X., Peng, J., Shen, L., Cao, Z., Xian, K., & Lin, G. (2025) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2025.3565642 · Cited by 3
“In most prior arts, dynamic NVS methods require RGBD data [43], [44] or multi-view videos [45], [46] as inputs.”
TE-NeRF: Triplane-Enhanced Neural Radiance Field for Artifact-Free Human Rendering
Mubashshira, S., & Desai, K. (2025) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw65960.2025.00030 · Cited by 3
“Recent learning-based methods [24, 33, 45] leverage human priors captured from training data to reconstruct 3D geometry and texture from monocular inputs.”
Liu, X., Yi, Z., Wu, X., & Shang, W. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02469-5 · Cited by 4
“For the reconstruction task, our method is superior to fusion-based method DynamicFusion Newcombe et al. (2015) and VolumeDeform Innmann et al. (2016) and exceeds the learning-based method DeepDeform Bozic et al. (2020b), NRT Bozic et al. (2020a), and OcclusionFusion Lin et al. (2022) in deformation error metric.”
BVI-CR: A Multi-View Human Dataset for Volumetric Video Compression
Gao, G., Azzarelli, A., Kwan, H. M., Anantrasirichai, N., Zhang, F., Andrew, W., Moolan-Feroze, O., & Bull, D. (2025) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas56072.2025.11044241 · Cited by 3
“The volumetric data generation is based on [28], and adapted for mutli-camera recording.”
CURL-SLAM: Continuous and Compact LiDAR Mapping
Zhang, K., Xu, S., Ding, Y., Kong, X., & Wang, S. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3588442 · Cited by 3
“To improve mapping efficiency, the weighted distance from each voxel to the nearest surfacealongthesensorrayisusedtoupdatetheTSDFmap[16], [17],[18],[19].”
Less is more: An effective method to extract object features for visual dynamic SLAM
Zhang, J., Yuan, L., Ran, T., Jia, J., Yang, S., & Tang, L. (2025) · Displays · doi:10.1016/j.displa.2025.103224 · Cited by 3
MonoFlex3D: Low-Latency Glasses-Free 3D Teleconference from a Single Camera
Zhang, C., Cuan, L., Yang, Y., Qiao, L., Qiao, X., & Huang, Y. (2025) · ACM Workshop on Mobile Immersive Computing, Networking, and Systems (ImmerCom) · doi:10.1145/3742889.3768345 · Cited by 3
“The core of MonoFlex3D is the single RGB camera capture, cleverly transforming depth-based multi-view volumetric fusion [21, 28, 29] into online animation driving with Gaussian representation.”
RIFTCast: A Template-Free End-to-End Multi-View Live Telepresence Framework and Benchmark
Zingsheim, D., Plack, M., Dröge, H., Pfeifer, J., Stotko, P., Hullin, M. B., & Klein, R. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754789 · Cited by 3
“To also handle and reconstruct free-form, non-rigid scenes, DynamicFusion [48] represented the scene in a canonical space and estimated a volumetric warp field for transforming to the current live frame.”
Split4D: Decomposed 4D Scene Reconstruction Without Video Segmentation
Hu, Y., Yang, Y., Lin, H., Wang, Y., Dong, J., Deng, Y., Zhu, X., Jia, F., Bao, H., Zhou, X., et al. (2025) · ACM Transactions on Graphics · doi:10.1145/3763343 · Cited by 7
“Early approaches, such as those using RGB-D cameras and non-rigid tracking [Dou et al. 2016; Newcombe et al. 2015; Orts-Escolano et al. 2016], build TSDF models in canonical space but were limited in robustness and scalability.”
3D Scene Reconstruction from Depth Camera Data
Zanuttigh, P., Marin, G., Mutto, C. D., Dominio, F., Minto, L., & Cortelazzo, G. M. (2016) · Springer eBooks · doi:10.1007/978-3-319-30973-6_7 · Cited by 3
Color-Introduced Frame-to-Model Registration for 3D Reconstruction
Li, F., Du, Y., & Liu, R. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-51814-5_10 · Cited by 3
Semi-automatic Initial Registration for the iRay System: A User Study
Xie, T., Islam, M. M., Lumsden, A. B., & Kakadiaris, I. A. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-60928-7_3 · Cited by 3
“In our system, ICP is applied to register the 3D point clouds because ICP has been proven to be very efficient in depth-related work [8-9].”
MethodFrom Scene Flow to Visual Odometry Through Local and Global Regularisation in Markov Random Fields
Scona, R., Matsuki, H., & Davison, A. J. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3150859 · Cited by 3
“While not directly related, techniques like DynamicFusion [19] and VolumeDeform [20] focus on reconstructing a single deforming foreground object which is tracked using sparser deformation graphs.”
MethodGNPM: Geometric-Aware Neural Parametric Models
Mohamed, M., & Agapito, L. (2022) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv57658.2022.00029 · Cited by 3
“Inspired by Dynamic Fusion [17], Bozic et al. [4] learn a globally consistent deformation graph while learning dense surface details via local MLPs.”
Huang, C., Yan, S., & Burgard, W. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3542693 · Cited by 5
“Other approaches leveraged optical flow for object motion tracking [28] or non-rigid tracking [29].”
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.