DTAM: Dense tracking and mapping in real-time
The DOI matches its Crossref record; 16 of the 17 references it cites were found in a registry or an index.
Abstract
DTAM is a system for real-time camera tracking and reconstruction which relies not on feature extraction but dense, every pixel methods. As a single hand-held RGB camera flies over a static scene, we estimate detailed textured depth maps at selected keyframes to produce a surface patchwork with millions of vertices. We use the hundreds of images available in a video stream to improve the quality of a simple photometric data term, and minimise a global spatially regularised energy functional in a novel non-convex optimisation framework. Interleaved, we track the camera's 6DOF motion precisely by frame-rate whole image alignment against the entire dense model. Our algorithms are highly parallelisable throughout and DTAM achieves real-time performance using current commodity GPU hardware. We demonstrate that a dense model permits superior tracking performance under rapid motion compared to a state of the art method using features; and also show the additional usefulness of the dense model for real-time scene interaction in a physics-enhanced augmented reality application.
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
“However, in our network, multi-scale projections not only help to overcome gradient locality during training (see [33], [47] for a broader discussion of this well-known issue), but also aid in online error correction at runtime.”
From Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery (2020) · cites this paper as [33]
In the citing paper
“SFMLearner [33] demonstrated the unsupervised learning of unscaled egomotion and depth from RGB imagery.”
From Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery (2020) · cites this paper as [33]
In the citing paper
“For KITTI, VIOLearner RGB outperforms approaches SFMLearner [33] and the network of Zhan et al. [30].”
From Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery (2020) · cites this paper as [33]
In the citing paper
“For AirSim, VIOLearner (no IMU) outperforms approaches SFMLearner [33] and ORB-SLAM2-S in translation error.”
From Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery (2020) · cites this paper as [33]
In the citing paper
“Compared to other vision-only approaches, for KITTI, VIOLearner (no IMU) outperforms approaches SFMLearner [33] and the network of Zhan et al. [30] and yet is outperformed by ORB-SLAM-2 which uses full bundle adjustment.”
From Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery (2020) · cites this paper as [33]
Citations
Cites
Nonlinear total variation based noise removal algorithms
Rudin, L. I., Osher, S., & Fatemi, E. (1992) · Physica D: Nonlinear Phenomena · doi:10.1016/0167-2789(92)90242-f · Cited by 15,704
Parallel Tracking and Mapping for Small AR Workspaces
Klein, G., & Murray, D. (2007) · 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2007.4538852 · Cited by 4,346
A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging
Chambolle, A., & Pock, T. (2011) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-010-0251-1 · Cited by 4,336
Lucas-Kanade 20 Years On: A Unifying Framework
Baker, S., & Matthews, I. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000011205.11775.fd · Cited by 3,155
Fast cost-volume filtering for visual correspondence and beyond
Rhemann, C., Hosni, A., Bleyer, M., Rother, C., & Gelautz, M. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995372 · Cited by 485
Live dense reconstruction with a single moving camera
Newcombe, R. A., & Davison, A. J. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5539794 · Cited by 366
Self-citationImproving the Agility of Keyframe-Based SLAM
Klein, G., & Murray, D. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-88688-4_59 · Cited by 335
Robust multi-sensor image alignment
Irani, M., & Anandan, P. (2002) · Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271) · doi:10.1109/iccv.1998.710832 · Cited by 220
Real-Time Dense Geometry from a Handheld Camera
Stühmer, J., Gumhold, S., & Cremers, D. (2010) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-15986-2_2 · Cited by 205
Some First-Order Algorithms for Total Variation Based Image Restoration
Aujol, J. (2009) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-009-0149-y · Cited by 151
Cited by
Items 901–1,000 of 1,004
A 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.17.1.94 · Cited by 75
Maier, D., & Bennewitz, M. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6386098 · Cited by 12
“[13] presented DTAM, a system for monocular camera-based tracking and dense reconstruction”
MethodA Multi-view Structured-Light System for Highly Accurate 3D Modeling
Ha, H., Oh, T.-H., & Kweon, I. S. (2015) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2015.21 · Cited by 12
“varying illumination environment), the brightness consistency assumption [8, 19] is not satisfied when refining the geometry with multi-view observations.”
BackgroundMethodCuFusion2: Accurate and Denoised Volumetric 3D Object Reconstruction Using Depth Cameras
Zhang, C. (2019) · IEEE Access · doi:10.1109/access.2019.2911119 · Cited by 11
“Dense reconstruction systems [21]–[23] have also been proposed to replace point cloud-based systems.”
BackgroundSimultaneous Localization and Mapping
Stachniss, C. (2017) · Springer eBooks · doi:10.1007/978-3-662-47094-7_49 · Cited by 13
Curved surface patches for rough terrain perception
Kanoulas, D. (2014) · Digital Repository Service (Northeastern University) · doi:10.17760/d20005088 · Cited by 12
Direct Visual Odometry using Bit-PlanesPreprint
Alismail, H., Browning, B., & Lucey, S. (2016) · arXiv (Cornell University) · doi:10.13140/rg.2.1.4529.9602 · Cited by 14
Simple change detection from mobile light field cameras
Dansereau, D. G., Williams, S. B., & Corke, P. I. (2016) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2015.12.008 · Cited by 11
“Indeed, in single-camera scenarios change detection generally requires joint estimation of the scene geometry and the camera’s motion, which can only be accomplished using sophisticated, iterative optimization methods [27, 28].”
MethodHI-SLAM2: Geometry-Aware Gaussian SLAM for Fast Monocular Scene Reconstruction
Zhang, W., Cheng, Q., Skuddis, D., Zeller, N., Cremers, D., & Haala, N. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3626627 · Cited by 61
“While dense SLAM approaches [18], [19], [20] attempt to address this limitation through per-pixel depth estimation, they remain susceptible to significant depth noise and struggle to achieve complete, accurate reconstructions.”
Global Visual and Semantic Observations for Outdoor Robot Localization
Li, J., Hu, S., Li, Q., Chen, J., Leung, V. C. M., & Song, H. (2020) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2020.3045263 · Cited by 11
Det-SLAM: A semantic visual SLAM for highly dynamic scenes using Detectron2
Eslamian, A., & Ahmadzadeh, M. R. (2022) · Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS) · doi:10.1109/icspis56952.2022.10043931 · Cited by 14
“DTAM [1] is one of these algorithms that produce dense maps.”
BackgroundMACARONS: Mapping and Coverage Anticipation with RGB Online Self-Supervision
Guédon, A., Monnier, T., Monasse, P., & Lepetit, V. (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/cvpr52729.2023.00097 · Cited by 34
“Traditional non-deep approaches are efficient methods developed for SLAM [19,53,54,78], which can further be augmented with neural networks [3, 42, 64].”
MethodYOD-SLAM: An Indoor Dynamic VSLAM Algorithm Based on the YOLOv8 Model and Depth Information
Li, Y., Wang, Y., Lu, L., & An, Q. (2024) · Electronics · doi:10.3390/electronics13183633 · Cited by 13
Optimized visual inertial SLAM for complex indoor dynamic scenes using RGB-D camera
Chen, J., Zhang, W., Wei, D., & Liu, X. (2025) · Measurement · doi:10.1016/j.measurement.2024.116615 · Cited by 11
SlamDunk: Affordable Real-Time RGB-D SLAM
Fioraio, N., & Stefano, L. D. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16178-5_28 · Cited by 11
“by GPGPU processing, to attain dense 3D reconstruction in real-time [19].”
MethodTowards consistent reconstructions of indoor spaces based on 6D RGB-D odometry and KinectFusion
Dong, H., Figueroa, N., & Saddik, A. E. (2014) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2014.6942798 · Cited by 12
“followed the same parallel tracking/mapping idea, but instead of used dense every pixel methods for camera pose tracking [9].”
MethodOskiper, T., Sizintsev, M., Branzoi, V., Samarasekera, S., & Kumar, R. (2013) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2013.6671782 · Cited by 14
Robust Real-Time Registration of RGB-D Images using Multi-Resolution Surfel Representations
Stückler, J., & Behnke, S. (2012) · Cited by 11
Sensor-Aware Recognition and Tracking for Wide-Area Augmented Reality on Mobile Phones
Chen, J., Cao, R., & Wang, Y. (2015) · Sensors · doi:10.3390/s151229847 · Cited by 15
“In [29], the authors presented a real-time camera tracking and reconstruction system relying on alignment of every pixel rather than feature extraction.”
BackgroundMonocular Depth from Small Motion Video Accelerated
Ham, C., Chang, M.-F., Lucey, S., & Singh, S. P. N. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00071 · Cited by 11
“[12] and Wendel et al.”
How to Improve CNN-Based 6-DoF Camera Pose Estimation
Seifi, S., & Tuytelaars, T. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00471 · Cited by 13
“Although various solutions for vSLAM have been proposed for different environments and applications, many of them share a similar structure and therefore similar limitations [15], [16], [17], [18]: They often lose track due to motion blur, high speed rotations, partial occlusions and presence of dynamic objects in the…”
BackgroundLIV-LAM: LiDAR and Visual Localization and Mapping
Radmanesh, R., Wang, Z., Chipade, V. S., Tsechpenakis, G., & Panagou, D. (2020) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc45564.2020.9148037 · Cited by 12
“Recently, considerable progress has been made in motion estimation and mapping by using a monocular camera using both direct and indirect feature-based methods [13], [8], [6].”
BackgroundORBBuf: A Robust Buffering Method for Remote Visual SLAM
Wang, Y.-P., Zou, Z., Wang, C., Dong, Y.-J., Qiao, L., & Manocha, D. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9635950 · Cited by 13
“addition to feature-based approaches, there are direct SLAM algorithms that use all raw pixels, including DTAM [26], LSD-SLAM [27], and DSO [16].”
MethodEdge-based Visual Odometry with Stereo Cameras using Multiple Oriented Quadtrees
Kim, C., Kim, J., & Kim, H. J. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341486 · Cited by 12
“…thanks to standard pipelines such as sparse point-based approaches [1]–[3], direct methods which find an optimal camera motion minimizing intensity residual between two images [4]–[6], and iterative closest points (ICP)-based algorithms [7]–[10] that align a pair of large point sets, e.g. 3-”
BackgroundGeometry-Constrained Scale Estimation for Monocular Visual Odometry
Zhang, H., Wang, X., Yin, X., Du, M., Liu, C., & Chen, Q. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3093771 · Cited by 13
Accurate and Robust Stereo Direct Visual Odometry for Agricultural Environment
Yu, T., Zhou, J., Wang, L., & Xiong, S. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561074 · Cited by 11
“Some notable systems such as DTAM [13], LSD-SLAM [14], SVO [15] have been developed to obtain accurate motion estimation and ease the computation cost.”
MethodLiODOM: Adaptive local mapping for robust LiDAR-only odometry
Garcia‐Fidalgo, E., Bonnin‐Pascual, F., & Ortiz, A. (2022) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2022.104226 · Cited by 14
Pan, C., Li, F., Pan, Y., Wang, Y., Soja, B. S., Li, Z., & Gao, J. (2023) · GPS Solutions · doi:10.1007/s10291-023-01586-3 · Cited by 13
“Whereaf-ter, the well-known open solutions such as MonoSLAM (Davison et al. 2007), DTAM (Newcombe et al. 2011), and ORB-SLAM2 (Mur-Artal and Tardós 2017) significantly promoted the research and application of visual SLAM.”
Consistency analysis and improvement for single-camera localization
Hesch, J. A., & Roumeliotis, S. I. (2012) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2012.6239190 · Cited by 11
“Numerous vision-based localization approaches have been presented in the literature, including methods based on the Extended Kalman Filter (EKF) [5], the Unscented Kalman Filter (UKF) [15], and Batch-least Squares (BLS) [10, 14].”
BackgroundMethodSampling-Based Multiview Reconstruction without Correspondences for 3D Edges
Teney, D., & Piater, J. (2012) · International Conference on 3D Imaging, Modeling, Processing, Visualization & Transmission · doi:10.1109/3dimpvt.2012.28 · Cited by 14
“[18] goes beyond precisely localizable features by tracking surface patches under photometric constraints to provide a dense reconstruction, but is based on frame-to-frame tracking.”
BackgroundLive RGB-D camera tracking for television production studios
Tykkälä, T., Comport, A. I., Kämäräinen, J., & Hartikainen, H. (2013) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2013.02.009 · Cited by 16
“DTAM applies the photometrical cost function for pose estimation, but also refines 3D structure by minimizing photometrical reprojection error [12].”
BackgroundReal-World Normal Map Capture for Nearly Flat Reflective Surfaces
Jacquet, B., Häne, C., Köser, K., & Pollefeys, M. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.94 · Cited by 13
“Our work should be seen as fitting in a major trend in computer vision and graphics that consists of developing techniques for capturing 3D models [19, 17] and measuring surface properties [6, 4] of surfaces to generate faithful copies of the real-world.”
BackgroundReal Time Complete Dense Depth Reconstruction for a Monocular Camera
Huang, X., Fan, L., Zhang, J., Wu, Q., & Yuan, C. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.89 · Cited by 14
“DTAM [10] and REMODE 978-1-5090-1437-8/16 $31.00 © 2016 IEEE DOI 10.1109/CVPRW.2016.89 674 [11] are typical examples of complete dense methods, however, the quality of recovered depth map in [10, 11] is tradeoff with completeness and very low when the depth map is complete.”
BackgroundMethodInfluentialBio-inspired visual self-localization in real world scenarios using Slow Feature Analysis
Metka, B., Franzius, M., & Bauer-Wersing, U. (2018) · PLoS ONE · doi:10.1371/journal.pone.0203994 · Cited by 11
“They make use of the whole image [18], which yields a dense 3d reconstruction, or only image regions with high gradients [19], which requires less computational resources and results in a semi-dense reconstruction of the environment.”
BackgroundMun, J.-H., Jeon, M., & Lee, B. (2019) · Sensors · doi:10.3390/s19112459 · Cited by 11
“The traditionally used structure from motion (SfM) method [5] rapidly constructs a scene structure while using ego-motion.”
MethodReal-Time Monocular Visual SLAM by Combining Points and Lines
Wei, X., Huang, J., & Ma, X. (2019) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2019.00026 · Cited by 11
“Existing monocular SLAM can be classified into two main categories with respect to the information which is used: feature-based methods [1, 2, 3, 4] and direct methods [5, 6, 7].”
BackgroundDense 3D surface reconstruction of large-scale streetscape from vehicle-borne imagery and LiDAR
Lin, X., Yang, B., Wang, F., Li, J., & Wang, X. (2020) · International Journal of Digital Earth · doi:10.1080/17538947.2020.1862318 · Cited by 12
“Representative studies are DSO (Engel, Koltun, and Cremers 2018), DTAM (Newcombe, Lovegrove, and Davison 2011), LSD-SLAM (Engel, Stückler, and Cremers 2015), etc. DSO presents a direct sparse visual odometry that combines a direct probabilistic model with consistent, joint optimization of all model parameters, but gre…”
BackgroundMulti-Camera Sensor Fusion for Visual Odometry using Deep Uncertainty Estimation
Kaygusuz, N., Méndez, O., & Bowden, R. (2021) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc48978.2021.9565079 · Cited by 15
Multi-view stereo for large-scale scene reconstruction with MRF-based depth inference
Sun, S., Xu, D., Wu, H., Ying, H., & Mou, Y. (2022) · Computers & Graphics · doi:10.1016/j.cag.2022.06.009 · Cited by 12
Real-Time Crater-Based Monocular 3-D Pose Tracking for Planetary Landing and Navigation
Liu, C., Guo, W., Hu, W., Chen, R., & Liu, J. (2022) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2022.3184660 · Cited by 11
“In this article, we set μ as 1×10-4, as used in [47].”
MethodAn evaluation of robust cost functions for RGB direct mapping
Concha, A., & Civera, J. (2015) · European Conference on Mobile Robots · doi:10.1109/ecmr.2015.7324174 · Cited by 11
“Abstract— The so-called direct SLAM methods have shown an impressive performance in estimating a dense 3D reconstruction from RGB sequences in real-time [1], [2], [3].”
BackgroundA moving observer in a three-dimensional world
Glennerster, A. (2016) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2015.0265 · Cited by 13
“Modern computer vision systems can carry out this process in real time giving rise to a dense reconstruction (figure 1b) and a highly reliable recovery of the camera pose [4,11].”
Research on obstacles avoidance technology For UAV based on improved PTAM algorithm
Bai, G., Xiang, X., Zhu, H., Yin, D., & Zhu, L. (2015) · IEEE International Conference on Progress in Informatics and Computing · doi:10.1109/pic.2015.7489907 · Cited by 11
“Among all these newly proposed methods, Dense Tracking and Mapping (DTAM) is an outstanding frame [10].”
MethodRegNet: Learning the Optimization of Direct Image-to-Image Pose RegistrationPreprint
Han, L., Ji, M., Fang, L., & Nießner, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.10212 · Cited by 13
“DTAM [24] and DSO [11] have demonstrated attractive results of joint estimating pose and depth by minimizing the photometric error using monocular cameras.”
BackgroundTowards Utilizing Deep Uncertainty In Traditional SLAM
Esfahani, M. A., Wu, K., Yuan, S., & Wang, H. (2019) · International Conference on Control and Automation · doi:10.1109/icca.2019.8899572 · Cited by 13
“In the past, researchers classified methods from sparse [4] to dense [5] and direct [6] to indirect [1] methods.”
BackgroundFu, Z., Jin, Z., Zhang, C., Dai, Y., Gao, X., Wang, Z., Li, L., Ding, G., Hu, H., Wang, P., et al. (2021) · International Journal of Medical Robotics and Computer Assisted Surgery · doi:10.1002/rcs.2274 · Cited by 14
Structerf-SLAM: Neural implicit representation SLAM for structural environments
Wang, H., Cao, Y., Wei, X., Shou, Y., Shen, L., Xu, Z., & Ren, K. (2024) · Computers & Graphics · doi:10.1016/j.cag.2024.103893 · Cited by 11
KN-SLAM: Keypoints and Neural Implicit Encoding SLAM
Wu, X., Liu, Z., Tian, Y., Liu, Z., & Chen, W. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3378264 · Cited by 13
“One of the early examples of such a representation was DTAM [9].”
BackgroundQue, H., Gao, H., Shan, W., Yang, X., & Zhao, R. J. (2024) · Sensors · doi:10.3390/s24227297 · Cited by 10
“New-combe et al. [17] proposed DTAM, a fully direct system that works with all the raw pixel information and estimate depth values based on photometric errors.”
G²-Mapping: General Gaussian Mapping for Monocular, RGB-D, and LiDAR-Inertial-Visual Systems
Chen, L., Hu, B., Wang, J., Bu, S., Wang, G., Han, P., & Chen, J. (2025) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2025.3541255 · Cited by 11
“Existing visual SLAM systems commonly rely on either discrete handcrafted feature extraction or deep learning embed-dings, following the framework outlined in [7] regarding mapping and tracking.”
Object Detection-Based Visual SLAM Optimization Method for Dynamic Scene
Deng, M., Hu, J., Wen, J., Zhang, X., & Jin, Q. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3552797 · Cited by 12
“Consequently, visual SLAM has garnered significant research interest in recent years, leading to the emergence of several noteworthy visual SLAM systems, including MonoSLAM [5], DTAM [6], LSD-SLAM [7], and ORB-SLAM2 [8].”
PLFG-SLAM: A visual SLAM for indoor weak-texture environments with point-line feature glue matching
Gao, R., Wu, S., Zhang, L., Pan, L., Zhang, G., Wang, H., & Wang, X. (2025) · Measurement · doi:10.1016/j.measurement.2025.118435 · Cited by 11
Robust Camera Tracking by Combining Color and Depth Measurements
Bylow, E., Olsson, C., & Kahl, F. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.692 · Cited by 10
“For smaller scenes, it has been shown that dense 3D reconstruction in real-time is possible by using an ordinary hand held camera [11, 6].”
BackgroundDirect Stereo Visual Odometry based on Lines
Holzmann, T., Fraundorfer, F., & Bischof, H. (2016) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0005715604740485 · Cited by 11
“For example, DTAM (Newcombe et al., 2011) tracks the pose of a monocular camera given a dense model of the scene, by minimizing the photometric error of the current image according to the whole model.”
BackgroundTuran, M., Abdullah, A., Jamiruddin, R., Araújo, H. J., Konukoğlu, E., & Sitti, M. (2017) · Cited by 9
Mixed reality system for industrial environment: an evaluation study
Abate, A. F., Nappi, M., Narducci, F., & Ricciardi, S. (2017) · CAAI Transactions on Intelligence Technology · doi:10.1049/trit.2017.0017 · Cited by 10
Robust and efficient edge-based visual odometry
Yan, F., Li, Z., & Zhou, Z. (2022) · Computational Visual Media · doi:10.1007/s41095-021-0251-7 · Cited by 10
“[35] presented dense tracking and mapping (DTAM), which achieves dense and smooth depth estimation by using a non-convex optimization process.”
BackgroundiMODE:Real-Time Incremental Monocular Dense Mapping Using Neural Field
Matsuki, H., Sucar, E., Laidow, T., Wada, K., Scona, R., & Davison, A. J. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10161538 · Cited by 10
“Rising processing power in particular made it feasible to develop real-time systems which built dense scene models [8], [9].”
BackgroundSelf-citationFast Monocular Scene Reconstruction with Global-Sparse Local-Dense Grids
Dong, W. F., Choy, C., Loop, C., Litany, O., Zhu, Y., & Anandkumar, A. (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/cvpr52729.2023.00415 · Cited by 13
“A variety of classical and learning-based methods [27, 38, 46, 50] have been developed to achieve high quality multi-view depth reconstruction from monocular images.”
BackgroundDeep Feature-based RGB-D Odometry using SuperPoint and SuperGlue
Fujimoto, S., & Matsunaga, N. (2023) · Procedia Computer Science · doi:10.1016/j.procs.2023.10.624 · Cited by 14
Physics-based Augmented Reality for 3D Deformable Object
Haouchine, N., Dequidt, J., Kerrien, E., Berger, M., & Cotin, S. (2012) · doi:10.2312/pe/vriphys/vriphys12/031-038 · Cited by 10
Leelasawassuk, T., & Mayol-Cuevas, W. W. (2013) · International Symposium on Wearable Computers · doi:10.1145/2493988.2494327 · Cited by 10
“We use the multi-view dense reconstruction convex optimization method [10, 30, 23], to get the depth of each referenced image.”
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.”
BackgroundTowards scheduling hard real-time image processing tasks on a single GPU
Golyanik, V., Nasri, M., & Stricker, D. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8297110 · Cited by 10
“Nowadays, Graphics Processing Units (GPU) are becoming indispensable hardware accelerators extensively used in image analysis applications with real-time constraints such as tracking and mapping [1], 3D reconstruction [2], angiography treatment [3], gesture recognition [4], advanced driver assistance systems [5] and s…”
BackgroundDirect visual-inertial odometry with semi-dense mapping
Xu, W., Choi, D., & Wang, G. (2018) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2018.02.002 · Cited by 11
“Visual odometry [18] estimates the depth of features, based on which, track the pose of the camera.”
Methods of Mobile Robot Visual Navigation and Environment Mapping
Pershina, J. S., Kazdorf, S. Y., & Лопота, А. В. (2019) · Optoelectronics Instrumentation and Data Processing · doi:10.3103/s8756699019020109 · Cited by 11
“Key algorithms in visual navigation based on the calculation of descriptors for local neighborhoods of point features are MonoSLAM [3], PTAM [8], and ORB-SLAM [4].”
BackgroundMethodInfluentialZeng, F., Jacobson, A., Smith, D., Boswell, N., Peynot, T., & Milford, M. J. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2932579 · Cited by 10
Combined Edge- and Stixel-based Object Detection in 3D Point Cloud
Hu, F., Yang, D., & Li, Y. (2019) · Sensors · doi:10.3390/s19204423 · Cited by 10
“With the evolution of 3D reconstruction techniques such as parallel tracking and mapping (PTAM) [23], dense tracking and mapping (DTAM) [24], and Kinect fusion [25], a potential solution is to exploit the merged point cloud, which can achieve high performance for indoor scenes where laser data are classified and modele…”
BackgroundFusing Semantics and Motion State Detection for Robust Visual SLAM
Singh, G., Wu, M., & Lam, S.-K. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093359 · Cited by 10
Modified Keyframe Selection Algorithm and Map Visualization Based on ORB-SLAM2
Xie, P., Su, W., Li, B., Jian, R., Huang, R., Zhang, S., & Wei, J. (2020) · International Conference on Robotics and Automation Sciences (ICRAS) · doi:10.1109/icras49812.2020.9135058 · Cited by 10
“DTAM [3], [4], DVO [5], [6], RGBD-SLAM [7], Kinect Fusion [8], [9] , ElaticFusion [10], [11] and ORB-SLAM2 [12] are famous RGBD-based algorithms.”
Mesoscopic photogrammetry with an unstabilized phone camera
Zhou, K. C., Cooke, C., Park, J., Qian, R., Horstmeyer, R. W., Izatt, J. A., & Farsiu, S. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00745 · Cited by 11
“In more recent years, there has been a resurgence of interest in direct approaches [39, 11, 9, 3, 10], including those incorporating deep learning.”
BackgroundMonocular Visual Inertial Direct SLAM with Robust Scale Estimation for Ground Robots/Vehicles
Sahoo, B., Biglarbegian, M., & Melek, W. (2021) · Robotics · doi:10.3390/robotics10010023 · Cited by 8
“The reader is encouraged to refer to Reference [16] for details. σ2d := α 2(σ2λ,photometric + σ 2 λ,geometric) (33) with; σ2λ,photometric = 2σ2i g2p (34) σ2λ,geometric = σ2l 〈g, l〉2 , (35) where; σi: camera-pixel noise σl : variance of positioning error of the initial point on epipolar line gp: gradient along the epip…”
BackgroundMethodInfluentialPath-Following Control Based on Ground-Watching Navigation
Miao, Q., & Cebon, D. (2017) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2017.2761772 · Cited by 11
“localizing ego vehicle on the map [9]–[12]; (iii).”
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
“The most classic reconstruction algorithms in SLAM based on dense point clouds are DTAM [26] and KinectFusion.”
MethodAhmadi, M., Naeini, A. A., Sheikholeslami, M. M., Arjmandi, Z., Zhang, Y., & Sohn, G. (2023) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case56687.2023.10260361 · Cited by 13
“However, direct methods like [15, 16], on the other hand, utilize the whole RGB-D image to minimize the photometric”
MethodA 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
Robust Arm and Hand Tracking by Unsupervised Context Learning
Spruyt, V., Ledda, A., & Philips, W. R. (2014) · Sensors · doi:10.3390/s140712023 · Cited by 12
“Although real-time, dense optical flow algorithms exist [27,28], in the context of our particle filter framework the optical flow estimation stage should only consume a fraction of the available CPU power.”
BackgroundSe, S., & Pears, N. E. (2012) · Springer eBooks · doi:10.1007/978-1-4471-4063-4_2 · Cited by 15
Visual odometry with a single-camera stereo omnidirectional system
Jaramillo, C., Yang, L., Muñoz, J. P., Taguchi, Y., & Xiao, J. (2019) · Machine Vision and Applications · doi:10.1007/s00138-019-01041-9 · Cited by 10
“Our purpose is not to design a SLAM framework since plenty of alternatives exist already, i.e., [4,5,6].”
BackgroundLocal optimized and scalable frame-to-model SLAM
Li, J., Wang, L., Yang, L., Zhang, J., Li, D., & Zhang, M. (2015) · Multimedia Tools and Applications · doi:10.1007/s11042-015-2780-5 · Cited by 10
“In recent years, dense tracking and mapping methods are used for taking full advantage of substantial parts of the acquired image data [3, 19, 22].”
InfluentialORB-LINE-SLAM: An Open-Source Stereo Visual SLAM System with Point and Line FeaturesPreprint
Alamanos, I., & Tzafestas, C. S. (2022) · doi:10.36227/techrxiv.21691949 · Cited by 11
Multi-Layer Scene Representation from Composed Focal Stacks
Ishikawa, R., Saitô, H., Kalkofen, D., & Mori, S. (2023) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2023.3320248 · Cited by 9
“With the increased multi-view inputs, we expect more stable multi-layer scene reconstruction, which may follow the same analogy to dense simultaneous localization and mapping (SLAM) [37] compared with keyframe-based SLAM [23].”
BackgroundMethodPark, G., Song, S., Sang, H., Im, D., Han, D., Kim, S., Lee, H., & Yoo, H. (2024) · IEEE transactions on circuits and systems for artificial intelligence. · doi:10.1109/tcasai.2024.3424388 · Cited by 9
“However, they rely on the feature extraction algorithm, such as features from accelerated segment test (FAST) which is vulnerable to texture-less regions, fast camera motion, etc. [9].”
MethodZhang, G., Tang, X., Wang, L., Cui, H., Fei, T., Tang, H., & Jiang, S. (2024) · Complex & Intelligent Systems · doi:10.1007/s40747-024-01575-0 · Cited by 12
“Depth estimation, as a fundamental task in the field of computer vision, has been widely used in areas such as autonomous driving [1], robot navigation [2, 3] and virtual reality [4].”
GraphAVO: Self-Supervised Visual Odometry Based on Graph-Assisted Geometric Consistency
Song, R., Liu, J., Xiao, Z., & Yan, B. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3462596 · Cited by 9
Schieber, H., Deuser, F., Egger, B., Oswald, N., & Roth, D. (2024) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2024.104206 · Cited by 9
Matiki, T. N., Narazaki, Y., Chowdhary, G. V., & Spencer, B. F. (2024) · Advances in Structural Engineering · doi:10.1177/13694332241291256 · Cited by 11
“…detection and mapping can be successful in controlled indoor settings and for simple geometry, local features such as corners and edges can suffer signi fi cant occlusions and misrepresentations, as have been highlighted by (Klein and Murray, 2007; Mur-Artal et al., 2015; Newcombe et al., 2011).”
EC-SLAM: Effectively constrained neural RGB-D SLAM with TSDF hash encoding and joint optimization
Li, G., Chen, Q., Yan, Y., & Pu, J. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112034 · Cited by 21
Improved ICP-based Pose Estimation by Distance-aware 3D Mapping
Hemmat, H. J., Bondarev, E., Dubbelman, G., & de With, P. H. N. (2014) · International Conference on Computer Vision Theory and Applications · doi:10.5220/0004716403600367 · Cited by 9
“These applications include KinectFusion (Newcombe et al., 2011a; Izadi et al., 2011), Kintinious (Whelan et al., 2012; Whe-lan et al., 2013), open source KinFu (PCL, 2011), and KinFu Large Scale (Bondarev et al., 2013).”
MethodInfluentialSDVL: Efficient and Accurate Semi-Direct Visual Localization
Perdices, E., & Cañas, J. (2019) · Italian National Conference on Sensors · doi:10.3390/s19020302 · Cited by 9
“Other authors improved efficiency and accuracy using direct methods in following years: Dense Tracking and Mapping (DTAM) [11], Large-Scale Direct Monocular SLAM (LSD-SLAM) [12] and in particular Direct Sparse Odometry (DSO) [13,14], which achieved real-time tracking by reducing map density.”
MethodMyasnikov, V., RAS, I. R. –. B. O. T. F. “. A. P., & Dmitriev, E. A. (2019) · Computer Optics · doi:10.18287/2412-6179-2019-43-3-492-503 · Cited by 9
Novel robust simultaneous localization and mapping for long-term autonomous robots
Wei, W., Zhu, X., & Wang, Y. (2022) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2000358 · Cited by 10
“For dense visual SLAM, dense tracking and mapping (DTAM) (Newcombe et al., 2011) is a technique involving a novel non-convex optimization framework that reconstructs a three-dimensional (3D) model in real time.”
MethodAccuracy evaluations of real-time LiDAR-based indoor localization system
Nimura, M., Kanai, K., & Katto, J. (2023) · International Conference on Consumer Electronics · doi:10.1109/icce56470.2023.10043573 · Cited by 13
“Another approach is called a direct method (e.g., DTAM [10], LSD-SLAM [11], DSO [12]).”
MethodLocalization of Mobile Robots Based on Depth Camera
Yin, Z., Wen, H., Nie, W., & Zhou, M. (2023) · Remote Sensing · doi:10.3390/rs15164016 · Cited by 9
“However, it requires GPU acceleration, increases costs, and is time-consuming [20].”
BackgroundScaling Up Edge-Assisted Real-Time Collaborative Visual SLAM Applications
Cao, H., Xu, J., Yang, Z., Shangguan, L., Zhang, J., He, X., & Liu, Y. (2023) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2023.3330763 · Cited by 16
“Typically, systems use monocular cameras [40], [41], stereo cameras [1], or RGB-D cameras [42].”
BackgroundA stereo vision SLAM with moving vehicles tracking in outdoor environment
Hong, C., Zhong, M., Jia, Z., You, C., & Wang, Z. (2023) · Machine Vision and Applications · doi:10.1007/s00138-023-01488-x · Cited by 9
SimpleMapping: Real-Time Visual-Inertial Dense Mapping with Deep Multi-View Stereo
Xin, Y., Zuo, X., Lu, D., & Leutenegger, S. (2023) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar59233.2023.00042 · Cited by 14
“Traditional real-time monocular visual(-inertial) SLAM systems [12,29,32,34,39,62] perform (semi-)dense mapping by matching pixel observations of 3D points and refining their depth values a r X i v : 2306 .”
BackgroundDesign and implementation of a mini quadrotor control system in GPS denied environments
Liu, C., & Prior, S. D. (2015) · International Conference on Unmanned Aircraft Systems (ICUAS) · doi:10.1109/icuas.2015.7152324 · Cited by 11
Hemmat, H. J., Bondarev, E., & de With, P. H. N. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-04114-8_35 · Cited by 9
“These types include the Signed Distance Function (SDF) model aggregating the complete set of sensor data in a temporal manner [1], the Octree-based model providing an optimum structure in terms of memory usage efficiency [13, 14], the straightforward and expeditious Points Cloud structure [15], the voxel-based [3, 7,…”
MethodMono-spectrum marker: an AR marker robust to image blur and defocus
Toyoura, M., Aruga, H., Turk, M., & Mao, X. (2013) · The Visual Computer · doi:10.1007/s00371-013-0910-5 · Cited by 9
“Systems like PTAM [10] or DTAM [13] for registration of real environment and virtual objects are constructed based on the map of natural features.”
Background
Metadata
- Authors
- Newcombe, R. A., Lovegrove, S. J., & Davison, A. J.
- Affiliations
- Imperial College London (GB)
- ORCID
- Davison, Andrew J.
- Venue
- 2011 International Conference on Computer Vision
- Publisher
- IEEE
- Published
- 2011-11
- Pages
- 2320-2327
- OpenAlex
- W2108134361
- Cited by
- 2,078
- References
- 17
- Citations by year
- 2026: 49 · 2025: 144 · 2024: 163 · 2023: 140 · 2022: 141 · 2021: 166 · 2020: 179 · 2019: 199 · 2018: 183 · 2017: 176 · 2016: 136 · 2015: 151 · 2014: 102 · 2013: 92 · 2012: 51
- FWCI
- 154.56
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Keywords
- dense 3D reconstruction, real-time mapping, depth images, non-convex optimization, energy minimization, image registration, GPU parallelization, augmented reality, handheld camera
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.