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 701–800 of 956
Shan, J., Zhang, Z., Li, H., Hsieh, C.-T., Li, Y., Wu, W., & Wang, H. (2025) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2025.3550269 · Cited by 4
EasyVis: a real-time 3D visualization software system for laparoscopic surgery box trainer
Sun, Y.-H., Ke, J., Fernandes, J., Chen, J., Jiang, H., & Hu, Y. H. (2025) · Updates in Surgery · doi:10.1007/s13304-025-02153-w · Cited by 3
Robust Structure and Motion Factorization of Non-Rigid Objects
Wang, G. (2015) · Frontiers in Robotics and AI · doi:10.3389/frobt.2015.00030 · Cited by 3
Global alignment of deformable objects captured by a single RGB-D camera
Guo, D., Li, K., Lai, Y., & Yang, J. (2017) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019318 · Cited by 3
“Alternative methods [1, 15] are also based on tracking and fusion of RGB-D sequences of non-rigidly deforming objects, although with different formulations.”
MethodPartial Single- and Multishape Dense Correspondence Using Functional Maps
Litany, O., Rodolà, E., Bronstein, A. M., Bronstein, M. M., & Cremers, D. (2018) · Handbook of numerical analysis · doi:10.1016/bs.hna.2018.09.002 · Cited by 3
“Recent works of [19,39] generalizing the Kinect fusion approach [40], were based on volumetric representation of 3D data.”
MethodContinuous Optimization Framework for Depth Sensor Viewpoint Selection
Maleki, B., Alempijevic, A., & Vidal‐Calleja, T. (2019) · Springer proceedings in advanced robotics · doi:10.1007/978-3-030-44051-0_21 · Cited by 3
“Systems that perform dense reconstruction [15] continuously add measurements into an underlying representation, the rapid dynamic nature of animal motion and inability to reasonably constrain animals for any periods of time renders these approaches unfeasible on a scale.”
BackgroundDepth texture synthesis for high-resolution reconstruction of large scenes
Labrie-Larrivée, F., Laurendeau, D., & Lalonde, J. (2019) · Machine Vision and Applications · doi:10.1007/s00138-019-01030-y · Cited by 3
Altantawy, D. A., Saleh, A. I., & Kishk, S. E. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.08.074 · Cited by 3
A Real-Time Indoor Scene Analysis Method Based on RGBD Stream
Wang, C., & Qi, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2944140 · Cited by 3
“Current camera tracking first determines pairwise point clouds according to sparse or dense features, such as color [50], normal [6], gradient [50], and curvature [5].”
BackgroundFragmentFusion: A Light-Weight SLAM Pipeline for Dense Reconstruction
Rückert, D., Innmann, M., & Stamminger, M. (2019) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct.2019.00-15 · Cited by 3
“Various publications presented extensions in terms of scale [15,20], the handling of dynamic or non-rigid scenes [7,13], and better global tracking with loop-closure [4].”
BackgroundInteracting with Real Objects in Virtual Worlds
Taylor, C., & Cosker, D. P. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-41816-8_15 · Cited by 3
Reconstruction of Colored Soft Deformable Objects Based on Self-Generated Template
Li, J., Liu, X., Deng, H., Wang, T., Lu, G., & Wang, J. (2021) · Computer-Aided Design · doi:10.1016/j.cad.2021.103124 · Cited by 3
Active SLAM in 3D deformable environments
Xu, M., Zhao, L., Huang, S., & Hao, Q. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9982224 · Cited by 3
“is used to observe the environment, a common approach is to deform the prior or build map directly based on the observations [1] [2] [3].”
MethodBreak and Splice: A Statistical Method for Non‐Rigid Point Cloud Registration
Gao, Q., Zhao, Y., Xi, L., Tang, W. J., & Wan, T. R. (2023) · Computer Graphics Forum · doi:10.1111/cgf.14788 · Cited by 3
Liu, C., Wang, A., Bu, C., Liu, T., Sui, J., & He, S. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3271002 · Cited by 3
“To obtain a measurement system that is portable and easy to use, many approaches have focused on obtaining a complete mesh by temporally fusing scans captured using a setup with a single red green blue-depth (RGB-D) camera [20], [21], [22], [23].”
MethodCan Robots Mold Soft Plastic Materials by Shaping Depth Images?
Gursoy, E., Tarbouriech, S., & Cherubini, A. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2023.3288836 · Cited by 3
Vo-Ho, V.-K., Pham, T.-T., Yamazaki, K., Tran, M., & Le, N. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00375 · Cited by 3
“DynamicFusion (20) is a seminal work in this approach, which aggregates observable object surface through each frame by optimizing a dense volumetric 6D motion field.”
BackgroundMethodInfluentialOnline 3D Reconstruction Based On Lidar Point Cloud
Han, Z., Fang, H., Yang, Q., Bai, Y., & Chen, L. (2023) · Chinese Control Conference (CCC) · doi:10.23919/ccc58697.2023.10240819 · Cited by 3
GeoAvatar: Geometrically-Consistent Multi-Person Avatar Reconstruction from Sparse Multi-View Videos
Lee, S., Kim, S., Lee, H.-K., Jeong, W.-S., & Lee, J. H. (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/cvpr52734.2025.01969 · Cited by 3
“Dynamic Avatar Reconstruction Early approaches to avatar reconstruction focus on simultaneous surface scanning and non-rigid tracking [4, 13, 30, 39].”
SMORE: Simultaneous Map and Object REconstruction
Chodosh, N., Madan, A., Lucey, S., & Ramanan, D. (2025) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv66043.2025.00046 · Cited by 5
“This problem has been widely studied in the context of handheld RGB-D sensors capturing human-scale scenes [23, 29, 44, 48].”
3D Shape Recovery of Deformable Soft-tissue with Computed Tomography and Depth Scan
Song, J., Wang, J., Zhao, L., Huang, S., & Dissanayake, G. (2016) · UTS ePRESS (University of Technology Sydney) · Cited by 3
ReMagicMirror: Action Learning Using Human Reenactment with the Mirror Metaphor
Dayrit, F. L., Kimura, R., Nakashima, Y., Blanco, A., Kawasaki, H., Ikeuchi, K., Sato, T., & Yokoya, N. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-51811-4_25 · Cited by 5
“This approach was later extended to handle non-rigid objects by describing the deformation of objects with transformations of signed distance field [9, 13, 14].”
MethodDevelopment of a Laboratory Framework for Testing Simultaneous Localization and Mapping Approaches
Neduchal, P., & Flídr, M. (2016) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2016.12.089 · Cited by 3
Direct-from-Video: Unsupervised NRSfM
Lebeda, K., Hadfield, S., & Bowden, R. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-49409-8_50 · Cited by 3
“Zollhofer Newcombe Garg Yu Proposed Property [15] [16] [17] [18]”
BackgroundDepth image super-resolution via semi self-taught learning framework
Zhao, F., Cao, Z., Xiao, Y., Zhang, X., Xian, K., & Li, R. (2017) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2270079 · Cited by 3
Space-Time Localization and Mapping
Lee, M., & Fowlkes, C. C. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.422 · Cited by 3
“When correspondence is not available but smooth surfaces are densely sampled in space-time, surface tracking approaches can be used to fuse observations (e.g., [26, 28]).”
MethodTemplate-Based 3D Reconstruction of Non-rigid Deformable Object from Monocular Video
Liu, Y., Peng, X., Zhou, W., Liu, B., & Gerndt, A. (2018) · 3D Research · doi:10.1007/s13319-018-0174-y · Cited by 3
Reconstructing non-rigid object with large movement using a single depth camera
Lu, F., Zhou, B., Lu, F., Zhang, Y., Chen, X., & Zhao, Q. (2018) · Computer Aided Geometric Design · doi:10.1016/j.cagd.2018.06.002 · Cited by 3
A Novel Kinect V2 Registration Method Using Color and Deep Geometry Descriptors
Gao, Y., Michels, T., & Koch, R. (2018) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553327 · Cited by 3
“The precise calibration of the two Kinect V2 sensors on this rig is critical to the dense 3D reconstruction of a large-scale and non-rigid scene [6], which can be further used for the novel view synthesis in the Free Viewpoint Video (FVV) [7] and Head-Mounted Display (HMD) [8] systems, together with the dynamic light…”
MethodBarrow, M., Burns, S. M., & Kastner, R. (2018) · International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl.2018.00064 · Cited by 3
“[1] avoid pre-defined priors by taking one image scan to be a canonical frame.”
BackgroundrxKinFu: Moving Volume KinectFusion for 3D Perception and Robotics
Kanoulas, D., Tsagarakis, N. G., & Vona, M. A. (2018) · IEEE-RAS International Conference on Humanoid Robots (Humanoids) · doi:10.1109/humanoids.2018.8625008 · Cited by 4
“For instance, BundleFusion [24] was developed towards improving the accuracy of the fusion, DynamicFusion [25] towards the reconstruction and tracking of non-rigid scenes, and ScemanticFusion [26] and DARNN [27] towards incorporating object semantics into the reconstruction and tracking.”
MethodAn End-to-End System for Real-Time Dynamic Point Cloud Visualization
Hofer, H., Seitner, F. H., & Gelautz, M. (2018) · International Conference on 3D Immersion (IC3D) · doi:10.1109/ic3d.2018.8657915 · Cited by 3
“formance limitations, researchers have introduced various assumptions, pre-trained models or movement constraints [8, 12, 16, 23], to reduce the required processing time without limiting the reconstruction quality.”
BackgroundSeptiana, A. (2020) · Dinar Jurnal Ekonomi dan Keuangan Islam · doi:10.21107/dinar.v5i1.4639 · Cited by 3
Zhang, H., Liu, B., Shen, C., Zhou, H., & Liu, S. (2019) · IEEE International Conference on Mechatronics and Automation (ICMA) · doi:10.1109/icma.2019.8816557 · Cited by 4
Sayo, A., Onizuka, H., Thomas, D., Nakashima, Y., Kawasaki, H., & Ikeuchi, K. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34879-3_18 · Cited by 4
“Non-prior methods can reconstruct any shape and the resolution of the reconstructed model varies depending on the resolution of the camera [6, 14, 21, 25, 26].”
BackgroundMethod3D Human Reconstruction from an Image for Mobile Telepresence Systems
Takeda, Y., Matsuda, A., & Rekimoto, J. (2020) · IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW) · doi:10.1109/vrw50115.2020.00237 · Cited by 3
“To reduce blind spots, therefore, a method is proposed for 3D-reconstructing a workspace using images showing the blind spot in a past frame by moving the depth camera and past shots [8].”
MethodEdge-Aware Monocular Dense Depth Estimation with Morphology
Li, Z., Zhu, X., Yu, H., Zhang, Q., & Jiang, Y. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412578 · Cited by 5
“There are several methods are presented, such as KinectFusion [11], ElasticFusion [12] and DynamicFusion [13].”
MethodReal-Time Dense Reconstruction of Indoor Scene
Niu, J., Hu, Q., Niu, Y. F., Zhang, T., & Jha, S. K. (2021) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2021.017418 · Cited by 3
“In recent years, some new RGB-D sensors (such as Kinect V1, Kinect V2, Realsense SR300, Realsense D415) are used for the 3D (three dimensional) reconstruction of indoor scenes, and new algorithms are successively produced (such as KinectFusion [6], DynamicFusion [7], ElasticFusion [8], Fusion4D [9], BundleFusion [10]).”
MethodAn observable time series based SLAM algorithm for deforming environmentPreprint
Song, J., Zhao, L., Huang, S., & Dissanayake, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.08563 · Cited by 3
“Specifically, ED based approach [15]–[18] is widely used to define the deformation of environment.”
MethodSmooth Deformation Field-based Mismatch Removal in Real-timePreprint
Zhou, H., & Jayender, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.08553 · Cited by 3
“To eliminate the impact of outliers, robust kernel functions can be applied [28].”
MethodLocalization and Tracking of User-Defined Points on Deformable Objects for Robotic ManipulationPreprint
Dittus, S., Alt, B., Hermann, A., Katić, D., Jäkel, R., & Fleischer, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.09067 · Cited by 3
“Most SotA approaches use either a low-resolution reference model of the deformable object [17] or none at all [16], [18].”
Echo-Reconstruction: Audio-Augmented 3D Scene ReconstructionPreprint
Wilson, J., Rewkowski, N., Lin, M. C., & Fuchs, H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.02405 · Cited by 3
Semi-supervised Learning for Human Pose Recognition with RGB-D Light-Model
Wang, X., Zhang, G., Yu, D., & Liu, D. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-48896-7_72 · Cited by 2
Free-Viewpoint Video of Outdoor Sports Using a Flying Camera
Hong, Z. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72630-9_11 · Cited by 2
IGAF: Incremental Guided Attention Fusion for Depth Super-Resolution
Tragakis, A., Kaul, C., Mitchell, K. J., Dai, H., Murray‐Smith, R., & Faccio, D. (2024) · Sensors · doi:10.3390/s25010024 · Cited by 2
Wang, J., Li, S., & Huang, K. (2025) · Sensors · doi:10.3390/s25103062 · Cited by 2
“In the reconstruction of dynamic scenes, Richard A, in his Dynamic-Fusion [21], pioneeringly uses volumetric 6D motion fields to represent the real-time deformation of objects, using multiple cameras to capture the dynamic actions of the human body; however, these studies are often used in VR and…”
Cin, A. P. D., Dikov, G., Ju, J., & Ghafoorian, M. (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/cvpr52734.2025.01554 · Cited by 3
“…model implements both unprojection and projection functions, enabling use in SfM. Motion in SfM is tackled by various methods [7, 9, 10, 25, 32, 36, 52, 63] that produce consistently scaled depth in dynamic regions, often leveraging monocular depths to align moving objects with static…”
Preconditioned Deformation Grids
Kaltheuner, J., Oebel, A., Dröge, H., Stotko, P., & Klein, R. (2025) · Computer Graphics Forum · doi:10.1111/cgf.70269 · Cited by 2
“This concept was extended in DynamicFusion [NFS15], which estimates a volumetric 6D warp field to incrementally align a canonical surface with each new input frame.”
FutureGS: Structured Gaussian Fields for Future-Aware Dynamic Scene Modeling
Ding, M., Wang, Z., Wang, J., Han, T., Hu, X., Ding, J., Tan, M., & Kuang, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755428 · Cited by 2
“Traditional methods address scene dynamics by estimating time-varying geometry [7, 29, 50].”
Gaussians on Fire: High-Frequency Reconstruction of Flames
Nazarenus, J., Michels, D. L., Pałubicki, W., Kou, S., Zhang, F.-L., Pirk, S., & Koch, R. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37152-2_25 · Cited by 2
“Before neural fields, non-rigid reconstruction often relied on RGB-D sensors [8, 21, 36, 64, 109] or strong low-rank and NRSfM priors [9, 16, 46, 66, 73].”
Autonomous robotic exploration of unknown soft objects
Hu, J., Jones, D., Galindo, G. L., Huang, S., & Valdastri, P. (2026) · The International Journal of Robotics Research · doi:10.1177/02783649251415415 · Cited by 2
“For dynamic reconstructions, methods aim to align sequential, temporally-spaced point clouds but face robustness issues in highly dynamic scenes (Newcombe et al., 2015).”
InfluentialAirGS: Real-Time 4D Gaussian Streaming for Free-Viewpoint Video Experiences
Wang, Z., Li, J., & Zhu, Y. (2026) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom59046.2026.11571182 · Cited by 2
“Existing dynamic scene reconstruction methods either explicitly model scene geometry using volumetric or mesh-based primitives [1], [2], or synthesize novel views through image-based interpolation [3]–[5].”
Vggt4d: Mining Motion Cues in Visual Geometry Transformers for 4D Scene Reconstruction
Hu, Y., Cheng, C., Yu, S., Guo, X., & Wang, H. (2026) · doi:10.1109/cvprf71909.2026.00044 · Cited by 13
“Early methods often rely on RGB-D sensors, which limit applicability in real-world scenes [15, 24].”
3D shape template generation from RGB-D images capturing a moving and deforming object
Takehara, H., Nakashima, Y., Sato, T., & Yokoya, N. (2016) · Electronic Imaging · doi:10.2352/issn.2470-1173.2016.21.3dipm-407 · Cited by 2
“Newcombe et al.[11] pre- sented real-time non-rigid 3D shape reconstruction with GPUaccelerated optimization.”
BackgroundMethodInfluentialSimultaneous 3D Tracking and Reconstruction of Multiple Moving Rigid Objects
Ozawa, T., Nakajima, Y., & Saitô, H. (2019) · Asia Pacific Workshop on Mixed and Augmented Reality (APMAR) · doi:10.1109/apmar.2019.8709158 · Cited by 2
“As one of SLAMs targeting dynamic scenes that includes non-rigid objects, DynamicFusion [10] can be cited.”
Visualization/AR/VR/MR Systems
Ikeuchi, K., Matsushita, Y., Sagawa, R., Kawasaki, H., Mukaigawa, Y., Furukawa, R., & Miyazaki, D. (2020) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-56577-0_9 · Cited by 2
Spatiotemporal Texture Reconstruction for Dynamic Objects Using a Single RGB‐D Camera
Kim, H., Kim, J., Nam, H., Park, J., & Lee, S. (2021) · Computer Graphics Forum · doi:10.1111/cgf.142652 · Cited by 2
“3D reconstruction methods using RGB-D images have been developed for static scenes [NIH ∗ 11, NZIS13, DNZ ∗ 17] and dynamic objects [NFS15,IZN ∗ 16,DKD ∗ 16].”
MethodWang, X., Boukhayma, A., Stephanie, P., Desjardin, É., Loscos, C., & Multon, F. (2022) · International Conference on Interactive Media, Smart Systems and Emerging Technologies (IMET) · doi:10.1109/imet54801.2022.9929833 · Cited by 2
“namicFusion [31]) to update the current mesh in combination with the SMPL parametric model to construct an inner human body.”
Ranka, D., Mehta, R., & Chopra, P. (2022) · International Conference on Advances in Computing, Communication Control and Networking (ICAC3N) · doi:10.1109/icac3n56670.2022.10074580 · Cited by 2
Fine Back Surfaces Oriented Human Reconstruction for Single RGB‐D Images
Fang, X., Yu, Q., He, J., Wang, L., & Liu, Z. (2023) · Computer Graphics Forum · doi:10.1111/cgf.14971 · Cited by 2
PCRT: Multi-branch Point Cloud Reconstruction from a Single Image with Transformers
Bai, J., Yang, Z., Guan, Y., & Yu, Q. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-8432-9_21 · Cited by 2
Bidirectional Patch-Based Correlations With Local Rigidity for Global Nonrigid Registration
Zhao, M., Yu, X., Hu, X., Wang, F., Xu, L., & Ji, X. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3447044 · Cited by 2
Autonomous Route Planning for UAV-Based 3D Reconstruction
Arnaldo, C. G., Moreno, F. P., Suárez, M. Z., & Jurado, R. D.-A. (2025) · doi:10.3390/engproc2025090078 · Cited by 2
“This method provides detailed and precise data, even in areas with dense vegetation or complex terrain [9].”
Link to the Past: Temporal Propagation for Fast 3D Human Reconstruction from Monocular Video
Marchellus, M., Noor, N., & Park, I. K. (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/cvpr52734.2025.00580 · Cited by 2
“Depth-based techniques [5, 24, 25] rely on specialized sensors, while FOF [6] achieves 30 FPS but lacks texture inference capability.”
Achieving Flexible 3D Reconstruction Volumes for RGB-D and RGB Camera Based Approaches
Mock, S., Lensing, P., & Broll, W. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46418-3_20 · Cited by 2
Computing Temporal Alignments of Human Motion Sequences in Wide Clothing Using Geodesic Patches
Shahu, A., Yang, J., Franco, J.-S., Hétroy, F., & Wuhrer, S. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.27 · Cited by 2
“[15] reconstruct a non-rigidly deforming surface while estimating a per frame volumetric warp field that transforms this surface into the live frame.”
BackgroundDepth Texture Synthesis for Realistic Architectural Modeling
Labrie-Larrivée, F., Laurendeau, D., & Lalonde, J. (2016) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv.2016.77 · Cited by 2
“We differentiate our method from LiDAR by exploiting inexpensive hardware to perform the reconstruction.”
MethodTMAGIC: A Model-Free 3D Tracker
Lebeda, K., Hadfield, S., & Bowden, R. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2675343 · Cited by 2
“, in particular Kinect Fusion [19] and Dynamic Fusion [20].”
BackgroundTowards Non-rigid Reconstruction - How to Adapt Rigid RGB-D Reconstruction to Non-rigid Movements?
Wasenmüller, O., Schenkenberger, B., & Stricker, D. (2017) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0006172402940299 · Cited by 2
“The first approach was DynamicFusion (Newcombe et al., 2015), which extends KinectFusion by a warping field to model the nonrigid deformation.”
MethodDifferential G-Buffer Rendering for Mediated Reality Applications
Schwandt, T., & Broll, W. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-60928-7_30 · Cited by 2
Multi-view Non-rigid Refinement and Normal Selection for High Quality 3D Reconstruction
Haque, S. M., & Govindu, V. M. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.261 · Cited by 2
“Of the numerous approaches, the methods of RGBD Mapping [10], Kinect Fusion [11] and DynamicFusion [15] are representative.”
MethodShape-aware spatio-temporal descriptors for interaction classification
Pirk, S., Diamanti, O., Thibert, B., Xu, D., & Guibas, L. J. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8297139 · Cited by 2
“To help with common occlusion problems [20], we couple RGBD sensing with a virtual reality (VR) feedback loop into a novel capture setup, and observe hand motions performed on virtual objects, extracted e.”
MethodRepresenting a Partially Observed Non-Rigid 3D Human Using Eigen-Texture and Eigen-Deformation
Kimura, R., Sayo, A., Dayrit, F. L., Nakashima, Y., Kawasaki, H., Blanco, A., & Ikeuchi, K. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545658 · Cited by 2
Real-Time RGBD Reconstruction Using Structural Constraint for Indoor AR
Wang, C., & Qi, Y. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00776-8_26 · Cited by 2
Measuring the Effects of Temporal Coherence in Depth Estimation for Dynamic Scenes
Tsekourakis, I., & Mordohai, P. (2019) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00346 · Cited by 2
“Even recent methods for scene flow on RGB-D sequences captured by depth cameras only present qualitative evaluation [35, 36, 37, 38, 39, 40].”
MethodMulti-granularity environment perception based on octree occupancy grid
ZHANG, G., Wu, B., Xu, Y., & Ye, Y.-D. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09302-w · Cited by 2
“As a result, the local features generated by monocular, stereo and infrared cameras are used in sparse [5], semi-dense [9] and dense mapping [23].”
BackgroundVisual SLAM Location Methods Based on Complex Scenes: A Review
Zhang, H., & Peng, J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-57881-7_43 · Cited by 2
Model-Based Dynamic Human Tracking and Reconstruction During Dynamic SLAM
Zhang, H., Zhang, T., & Zhang, L. (2020) · Courses and lectures · doi:10.1007/978-3-030-58380-4_2 · Cited by 2
Efficient two step optimization for large embedded deformation graph based SLAM
Song, J., Bai, F., Zhao, L., Huang, S., & Xiong, R. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196930 · Cited by 2
“ED graph has been widely applied in RGBD and stereo SLAM [9] [10] and 3D human reconstructions [11] [12] (discrete hierarchical warp grid but very similar to ED graph) [13] [14] [15].”
BackgroundMethodInfluentialDeep Surface Normal Estimation on the 2-Sphere with Confidence Guided Semantic Attention
Li, Q., Guo, J. J., Fei, Y., Tang, Q., Sun, W., Zeng, J., & Guo, Y. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58586-0_43 · Cited by 2
“Recently, the availability of depth information has promoted a great enhancement in the applications of object recognition [10, 31], sematic segmentation [4, 13, 20, 28], 3D scene reconstruction [17, 22, 25], pose estimation [5, 37], etc .”
BackgroundMethodCapturing detailed deformations of moving human bodies
Chen, H., Park, H., Macit, K., & Kavan, L. (2021) · ACM Transactions on Graphics · doi:10.1145/3476576.3476642 · Cited by 3
Some Research Questions for SLAM in Deformable Environments
Huang, S., Chen, Y., Zhao, L., Zhang, Y., & Xu, M. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9635883 · Cited by 2
“For example, DynamicFusion algorithm [2] and SurfelWarp [3] apply dual quaternions [4] to represent the warp field, and provide real-time reconstruction and tracking systems for non-rigid scenes, one using volumetric data structure and the other using a surfel based representation of the geometry.”
MethodRGBD Infant Head Reconstruction for Cranial Vault Asymmetry Estimation
Zeitvogel, S., Wernet, C., Wetzel, J., Laubenheimer, A., & Stoevesandt, K. (2022) · IEEE Access · doi:10.1109/access.2022.3160749 · Cited by 2
Ground Enhanced RGB-D SLAM for Dynamic Environments
Guo, R., & Liu, X. (2021) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio54168.2021.9739362 · Cited by 2
Reconstructing Detailed Human Body Using a Viewpoint Auto-Changing RGB-D Sensor for Rescue Operation
Liu, C., Wang, A., Bu, C., Wang, W., Fang, Z., & He, S. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3178082 · Cited by 2
Park, B.-S., Lee, S., Park, J.-T., Kim, J.-K., Kim, W., & Seo, Y.-H. (2022) · Sensors · doi:10.3390/s22228815 · Cited by 2
“DynamicFusion [19] is a technology for the real-time reconstruction of a 3D model using a depth image captured by a single depth sensor, and the depth information acquired by a single RGB-Depth camera is gradually accumulated.”
BackgroundSPaM: soft patch matching for non-rigid pointcloud registration
Maleki, B., Falque, R., Vidal‐Calleja, T., & Alempijevic, A. (2023) · Frontiers in Robotics and AI · doi:10.3389/frobt.2023.1019579 · Cited by 2
“In contrast to existing dense SLAM approaches, the recent work SurfelWarp (Gao and Tedrake, 2019) enhances DynamicFusion by replacing volumetric data structures with surfel-based representation of geometry and a deformation field.”
MethodInfluentialExtracting Contact and Motion from Manipulation VideosPreprint
Zampogiannis, K., Ganguly, K., Fermuller, C. M., & Aloimonos, Y. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1807.04870 · Cited by 2
“The warp field estimation is performed either by iterating between correspondence estimation and motion optimization [22, 23, 24, 25], or in a correspondence-free fashion, by aligning volumetric SDFs (Signed Distance Fields) [26].”
BackgroundMethodInfluentialS4-Net: Geometry-Consistent Semi-Supervised Semantic SegmentationPreprint
Stekovic, S., Fraundorfer, F., & Lepetit, V. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.10717 · Cited by 2
“Alternatively, automated dense SLAM systems could be used to acquire such information for rigid scenes with a color camera [11, 36, 5], or for deformables scenes with an RGB-D camera [24, 34].”
MethodHead Reconstruction from Internet PhotosPreprint
Liang, S. Y., Shapiro, L. G., & Kemelmacher-Shlizerman, I. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.04763 · Cited by 2
“Examples include non rigid structure from motion methods [7,8], multiview methods [9,10], dynamic kinect fusion [11], and RGB-D based methods [12,13].”
MethodReal-time Simultaneous 3D Head Modeling and Facial Motion Capture with an RGB-D cameraPreprint
Thomas, D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.10557 · Cited by 2
“More recently, Newcombe et al. [8] extended KinectFusion to DynamicFusion, which allows capturing dynamic changes in the volumetric TSDF in real-time by using embedded deformation [17].”
BackgroundMethodResultInfluentialReconstruction of People in Loose ClothingPreprint
Jinka, S. S., Chacko, R., Srivastava, A., Sharma, A. K., & Narayanan, P. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.04778 · Cited by 2
“Advancements in commercial depth sensing over the last decade [38, 5, 47, 30] has partly helped in overcoming this limitation.”
Shape control of deformable objects by adaptive visual servoing
Lagneau, R. (2020) · theses.fr (ABES) · doi:10.70675/ae55b79az0c0ez48c2z8798z48ad095a7702 · Cited by 3
Non-Rigid Structure-from-Motion and Shading
Gallardo, M., Collins, T., & Bartoli, A. (2020) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-51866-0_4 · Cited by 1
“As all datasets, except the Kinect paper dataset, are poorly-textured, the correspondences are sparse.”
BackgroundMethodInfluential3D Human Shape Estimation from a Silhouette Image by using Statistical Human Shape Spaces
Ahn, D., & Park, S. I. (2023) · Journal of the Korea Computer Graphics Society · doi:10.15701/kcgs.2022.29.1.13 · Cited by 1
Modeling Detailed Human Geometry with Adaptive Local Refinement
Du, B., Chen, K., Zhang, H., Yin, F., Wu, B., & Nguyen, T. Q. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00571 · Cited by 1
“Traditional 3D-scan [27, 38] and multi-view reconstruction methods [14, 16, 23] rely heavily on hardware to capture dense inputs of a target scene, limiting its application in daily scenarios.”
BackgroundFang, L. (2024) · Advances in computer vision and pattern recognition · doi:10.1007/978-981-97-6915-5_4 · Cited by 1
Generalizable Geometry-Aware Human Radiance Modeling from Multi-view Images
Wu, W., Mo, Z., Yu, W., Cheng, Y., Zhang, T., & Huang, J. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-8508-7_7 · Cited by 1
Tang, Q., Yin, Y., Zhu, Y., Zhou, Y., Kuang, Z., Jianping Fan, B., Ding, J., & He, J. (2026) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2026.104815 · Cited by 1
ERNet: Efficient Non-Rigid Registration Network for Point Sequences
He, G., Xiao, Y., Xu, Z., Zhou, X., & Peng, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02521 · Cited by 1
“While many follow-up works [1, 7, 15, 33, 34] have introduced regularizers, such as As-Rigid-As-Possible constraint [26, 33] and deformation graph [1, 37], to improve the robustness of their optimization process, they still easily get stuck in local optima due to the non-convexity of the objective…”
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.