The DOI matches its Crossref record; 65 of the 71 references it cites were found in a registry or an index.
Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments
Abstract
We introduce a new dataset, Human3.6M, of 3.6 Million accurate 3D Human poses, acquired by recording the performance of 5 female and 6 male subjects, under 4 different viewpoints, for training realistic human sensing systems and for evaluating the next generation of human pose estimation models and algorithms. Besides increasing the size of the datasets in the current state-of-the-art by several orders of magnitude, we also aim to complement such datasets with a diverse set of motions and poses encountered as part of typical human activities (taking photos, talking on the phone, posing, greeting, eating, etc.), with additional synchronized image, human motion capture, and time of flight (depth) data, and with accurate 3D body scans of all the subject actors involved. We also provide controlled mixed reality evaluation scenarios where 3D human models are animated using motion capture and inserted using correct 3D geometry, in complex real environments, viewed with moving cameras, and under occlusion. Finally, we provide a set of large-scale statistical models and detailed evaluation baselines for the dataset illustrating its diversity and the scope for improvement by future work in the research community. Our experiments show that our best large-scale model can leverage our full training set to obtain a 20% improvement in performance compared to a training set of the scale of the largest existing public dataset for this problem. Yet the potential for improvement by leveraging higher capacity, more complex models with our large dataset, is substantially vaster and should stimulate future research. The dataset together with code for the associated large-scale learning models, features, visualization tools, as well as the evaluation server, is available online at http://vision.imar.ro/human3.6m.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Multiple View Geometry in Computer Vision
Hartley, R., & Zisserman, A. (2004) · doi:10.1017/cbo9780511811685 · Cited by 21,070
Nonlinear Component Analysis as a Kernel Eigenvalue Problem
Schölkopf, B., Smola, A. J., & Müller, K. (1998) · Neural Computation · doi:10.1162/089976698300017467 · Cited by 8,196
Real-time human pose recognition in parts from single depth images
Shotton, J., Fitzgibbon, A. W., Cook, M., Sharp, T., Finocchio, M., Moore, R., Kipman, A., & Blake, A. (2011) · 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.2011.5995316 · Cited by 3,545
Multiple View Geometry in Computer Vision
(2001) · Kybernetes · doi:10.1108/k.2001.30.9_10.1333.2 · Cited by 3,376
Random Features for Large-Scale Kernel Machines
Rahimi, A., & Recht, B. (2007) · Cited by 2,783
Kernel methods in machine learning
Hofmann, T., Schölkopf, B., & Smola, A. J. (2008) · The Annals of Statistics · doi:10.1214/009053607000000677 · Cited by 2,292
SCAPE: shape completion and animation of people
Anguelov, D., Srinivasan, P., Koller, D., Thrun, S., Rodgers, J., & Davis, J. (2005) · ACM Transactions on Graphics · doi:10.1145/1073204.1073207 · Cited by 1,489
KERNEL METHODS IN MACHINE LEARNING
Hofmann, T., Schölkopf, B., & Smola, A. J. (2008) · Cited by 1,408
Sigal, L., Bălan, A. O., & Black, M. J. (2009) · International Journal of Computer Vision · doi:10.1007/s11263-009-0273-6 · Cited by 1,205
Poselets: Body part detectors trained using 3D human pose annotations
Bourdev, L., & Malik, J. (2009) · 2009 IEEE 12th International Conference on Computer Vision · doi:10.1109/iccv.2009.5459303 · Cited by 961
Flagged first, then by citations
6 of the 71 references couldn’t be matched to a known paper.
Research works printed but not linked: 3
- 0, INRIA 4D
- 0, CMU HMC
- sminchisescu, 2011, Feature-based Human Pose Estimation in Guide to Visual Analysis of Humans Looking at People
Cited by
Sorted by citations
Items 701–800 of 1,009
The reliability and validity of gait analysis system using 3D markerless pose estimation algorithms
Liang, S., Zhang, Y., Diao, Y., Li, G., & Zhao, G. (2022) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2022.857975 · Cited by 38
Learning Constrained Dynamic Correlations in Spatiotemporal Graphs for Motion Prediction
Fu, J., Yang, F., Dang, Y., Liu, X., & Yin, J. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3277476 · Cited by 37
“6M [45], Carnegie Mel-lon University (CMU) Mocap, and 3D Poses in the Wild (3DPW) [46] datasets.”
Background3D-Aware Neural Body Fitting for Occlusion Robust 3D Human Pose Estimation
Zhang, Y., Ji, P., Wang, A., Mei, J., Kortylewski, A., & Yuille, A. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00862 · Cited by 44
Cai, Q., Hu, X., Hou, S., Yao, L., & Huang, Y. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i2.27847 · Cited by 41
3D pose estimation dataset and deep learning-based ergonomic risk assessment in construction
Fan, C., Mei, Q., & Li, X. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105452 · Cited by 37
3D Human Pose Tracking Priors using Geodesic Mixture Models
Simo‐Serra, E., Carme Torras, A., & Moreno-Noguer, F. (2016) · International Journal of Computer Vision · doi:10.1007/s11263-016-0941-2 · Cited by 37
“, the relative rotation between two consecutive joints [26].”
BackgroundKinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation
Kundu, J. N., Seth, S., Rahul, M. V., Rakesh, M., Babu, R. V., & Chakraborty, A. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.6792 · Cited by 37
“Available motion capture systems, such as body-worn sensors (IMUs) or multi-camera structure-from-motion (SFM), requires careful pre-calibration, and hence usually done in a pre-setup laboratory environment (Ionescu et al. 2013; Zhang et al. 2017).”
Hwang, D.-H., Aso, K., Yuan, Y., Kitani, K., & Koike, H. (2020) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3379337.3415856 · Cited by 37
“Human pose estimation with a single monocular RGB camera has developed rapidly with improvements in deep neural networks and large scale training datasets [4, 25, 58, 39].”
BackgroundWang, H., Xie, Z., Lu, L., Li, L., & Xu, X. (2021) · Journal of Biomechanics · doi:10.1016/j.jbiomech.2021.110860 · Cited by 36
RPM 2.0: RF-Based Pose Machines for Multi-Person 3D Pose Estimation
Xie, C., Zhang, D., Wu, Z., Yu, C., Hu, Y., & Chen, Y. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3287329 · Cited by 36
“We use the absolute Mean Per Joint Position Error (MPJPE) [67] to evaluate the performance of RPM 2.0 and other baseline methods, which measures the L 2 distances between the ground truth joints and the predicted joints in the world coordinate.”
MethodInfluentialLi, J., Wang, Z., Wang, C., & Su, W. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111810 · Cited by 38
Appearance Consensus Driven Self-supervised Human Mesh Recovery
Kundu, J. N., Rakesh, M., Jampani, V., Venkatesh, R., & Babu, R. V. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58452-8_46 · Cited by 38
GraphMDN: Leveraging graph structure and deep learning to solve inverse problems
Oikarinen, T., Hannah, D., & Kazerounian, S. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9534301 · Cited by 37
“6M dataset [14], which contains 3.6 million images attained by a motion capture system with four synchronized cameras in an indoor environment.”
InfluentialVonstad, E. K., Su, X., Vereijken, B., Bach, K., & Nilsen, J. H. (2020) · Sensors · doi:10.3390/s20236940 · Cited by 35
“6M dataset [40].”
LEMMA: A Multi-view Dataset for L Earning Multi-agent Multi-task Activities
Jia, B., Chen, Y., Huang, S., Zhu, Y., & Zhu, S. C. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58574-7_46 · Cited by 36
Learning Complex 3D Human Self-Contact
Fieraru, M., Zanfir, M., Oneaţă, E., Popa, A.-I., Olaru, V., & Sminchisescu, C. (2021) · AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) · doi:10.1609/aaai.v35i2.16223 · Cited by 41
Self-citationDeep learning pipelines for recognition of gait biometrics with covariates: a comprehensive review
Parashar, A., Parashar, A., Ding, W., Shekhawat, R. S., & Rida, I. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-022-10365-4 · Cited by 36
Cai, Z., Zhang, M., Ren, J., Wei, C., Ren, D., Lin, Z., Zhao, H., Yang, L., Loy, C. C., & Liu, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3450537 · Cited by 77
“6M [8] is fully indoor.”
BackgroundMethodInfluentialSárándi, I., Linder, T., Arras, K. O., & Leibe, B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.04987 · Cited by 41
“patches over some of the body joints. We have recently found that such techniques are also very eective for 3D pose estimation [20]. 3 Approach The dataset in this challenge is a subset of Human3.6M [9][8] with a few important dierences compared to the full benchmark. First, it lacks person bounding boxes and camera i…”
BackgroundGenerative Models as a Data Source for Multiview Representation LearningPreprint
Jahanian, A., Puig, X., Tian, Y., & Isola, P. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.05258 · Cited by 92
“…a source for training models, for example in semantic segmentation (Chen et al., 2019; Ros et al., 2016), human pose estimation (Varol et al., 2017; Ionescu et al., 2013; Shakhnarovich et al., 2003), optical flow (Mayer et al., 2016; Fischer et al., 2015), or self-supervised multitask learning (Ren…”
BackgroundMethodBidirectional Transformer GAN for Long-term Human Motion Prediction
Zhao, M., Tang, H., Xie, P., Dai, S., Sebe, N., & Wang, W. (2023) · ACM Transactions on Multimedia Computing, Communications, and Applications · doi:10.1145/3579359 · Cited by 38
“In particular, as shown in Table 2, for the MPJPE metric, our method exceeds STSGCN [39] and His.”
BackgroundMethodInfluentialA Systematic Review of Recent Deep Learning Approaches for 3D Human Pose Estimation
El Kaid, A., & Baïna, K. (2023) · Journal of Imaging · doi:10.3390/jimaging9120275 · Cited by 44
“The same authors proposed a method in [82] to learn a regression model for 3D pose mapping from videos using CNNs.”
MethodOut-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation
Guan, S., Xu, J., He, M. Z., Wang, Y., Ni, B., & Yang, X. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3194167 · Cited by 39
“6M [24] dataset used in our work contains about 3.”
BackgroundMethodFrom Human Pose Similarity Metric to 3D Human Pose Estimator: Temporal Propagating LSTM Networks
Lee, K., Kim, W., & Lee, S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3164344 · Cited by 36
“The metric scores were obtained using the most popular method, Mean Per Joint Position Error (MPJPE) [9].”
BackgroundMethodInfluentialQuantification of Occlusion Handling Capability of a 3D Human Pose Estimation Framework
Ghafoor, M., & Mahmood, A. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3158068 · Cited by 35
“6M [20], NTU RGB+D [21] and SYSU [22] and compared with five existing state of the art methods.”
BackgroundResultWeakly-supervised pre-training for 3D human pose estimation via perspective knowledge
Qiu, Z., Qiu, K., Fu, J., & Fu, D. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109497 · Cited by 35
Skeleton-based automatic assessment and prediction of intrusion risk in construction hazardous areas
Huang, H., Hu, H., Xu, F., Zhang, Z., & Tao, Y. (2023) · Safety Science · doi:10.1016/j.ssci.2023.106150 · Cited by 35
Deformable Mesh Transformer for 3D Human Mesh Recovery
Yoshiyasu, Y. (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/cvpr52729.2023.01631 · Cited by 35
“6M [21], MuCo-3DHP [34], UP3D [27], COCO [30], MPII [8], following [15, 28, 29, 35].”
Zolly: Zoom Focal Length Correctly for Perspective-Distorted Human Mesh Reconstruction
Wang, W., Ge, Y., Mei, H., Cai, Z., Sun, Q., Wang, Y., Shen, C., Yang, L., & Komura, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00363 · Cited by 35
“6M [8] dataset.”
MoConVQ: Unified Physics-Based Motion Control via Scalable Discrete Representations
Yao, H., Song, Z., Zhou, Y., Ao, T., Chen, B., & Liu, L. (2024) · ACM Transactions on Graphics · doi:10.1145/3658137 · Cited by 46
“6M dataset [Ionescu et al. 2014] and use our MoConVQ to track the results.”
Variational Autoencoded Regression: High Dimensional Regression of Visual Data on Complex Manifold
Yoo, Y., Yun, S., Chang, H. J., Demiris, Y., & Choi, J. Y. (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.2017.314 · Cited by 35
“6m) [20] dataset for generating proper human appearance given the joint positions.”
BackgroundAutomation of the Timed-Up-and-Go Test Using a Conventional Video Camera
Savoie, P., Cameron, J. A. D., Kaye, M. E., & Scheme, E. (2019) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2019.2934342 · Cited by 35
“6M [45] and Leeds Sports Pose [46] These datasets contain a combined amount of more than 3.”
BackgroundMethodLearning Multi-human Optical Flow
Ranjan, A., Hoffmann, D. T., Tzionas, D., Tang, S., Romero, J., & Black, M. J. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01279-w · Cited by 34
Scheck, T., Seidel, R., & Hirtz, G. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093563 · Cited by 35
HEMlets PoSh: Learning Part-Centric Heatmap Triplets for 3D Human Pose and Shape Estimation
Zhou, K., Han, X., Jiang, N., Jia, K., & Lu, J. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3051173 · Cited by 35
SMPLR: Deep learning based SMPL reverse for 3D human pose and shape recovery
Madadi, M., Bertiche, H., & Escalera, S. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107472 · Cited by 54
Learning to Estimate 3D Human Pose From Point Cloud
Zhou, Y., Dong, H., & Saddik, A. E. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.2999849 · Cited by 35
“Others approaches [4]–[6] regressed the heat maps for 3D pose estimation.”
BackgroundPoseNet3D: Learning Temporally Consistent 3D Human Pose via Knowledge Distillation
Tripathi, S., Ranade, S., Tyagi, A., & Agrawal, A. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00041 · Cited by 35
Human Pose and Shape Estimation From Single Polarization Images
Zou, S., Zuo, X., Wang, S., Qian, Y., Guo, C., & Cheng, L. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3162469 · Cited by 35
“extended PHSPD dataset is the largest to date against other 3D human pose estimation datasets [32], [33], and its multi-modality property renders its great potential in facilitating existing and new research directions.”
BackgroundUncertainty-Aware 3D Human Pose Estimation from Monocular Video
Zhang, J., Chen, Y., & Tu, Z. (2022) · ACM Multimedia · doi:10.1145/3503161.3547773 · Cited by 46
FLEX: Extrinsic Parameters-free Multi-view 3D Human Motion Reconstruction
Gordon, B., Raab, S., Azov, G., Giryes, R., & Cohen‐Or, D. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19827-4_11 · Cited by 36
“6M [29,9], CMU Panoptic [17] and TUM Shelf & Campus [3] contain static scenes only, while Tagging [79] and Ski-Pose PTZ-Camera [62] contain rotating cameras whose locations are fixed.”
BackgroundMethodInfluentialA Unified Framework for Domain Adaptive Pose Estimation
Kim, D., Wang, K., Saenko, K., Betke, M., & Sclaroff, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19827-4_35 · Cited by 36
Deep Learning for Vision-based Prediction: A SurveyPreprint
Rasouli, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.00095 · Cited by 51
“6M [237] Activities RGB, 3D Pose, Activity x x x MPII Human Pose [281] Activities RGB, Pose, Activity x Online RGBD Action Dataset (ORGBD) [282] Activities RGBD, BB, 3D Pose, Activity x Sports-1M [283] Sports RGB, Activity x x”
BackgroundDeep Learning Methods for 3D Human Pose Estimation under Different Supervision Paradigms: A Survey
Zhang, D., Wu, Y., Guo, M., & Chen, Y. (2021) · Electronics · doi:10.3390/electronics10182267 · Cited by 34
UV Volumes for Real-time Rendering of Editable Free-view Human Performance
Chen, Y., Wang, X., Chen, X., Zhang, Q., Li, X., Guo, Y., Wang, J., & Wang, F. (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/cvpr52729.2023.01595 · Cited by 34
Being Comes from Not-Being: Open-Vocabulary Text-to-Motion Generation with Wordless Training
Lin, J., Chang, J., Liu, L., Li, G., Lin, L., Tian, Q. C., & Chen, C. W. (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/cvpr52729.2023.02224 · Cited by 43
“The success of these methods is heavily dependent on large motion capture datasets [6, 7, 14, 20, 21, 43, 47] and labeled motion description datasets, including AMASS [34], KIT motion-language dataset [32], and HumanML3D dataset [10].”
MethodGlobal and Local Spatio-Temporal Encoder for 3D Human Pose Estimation
Wang, Y., Kang, H., Wu, D., Yang, W., & Zhang, L. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3321438 · Cited by 34
“6M [47] and MPI-INF-3DHP [48].”
EgoHumans: An Egocentric 3D Multi-Human Benchmark
Khirodkar, R., Bansal, A., Ma, L., Newcombe, R., Vo, M., & Kitani, K. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01814 · Cited by 44
“As a result, existing popular 3D benchmarks [37, 43, 48, 66, 82, 110] are constrained to indoor environments or, at most, two human subjects if outdoors, stationary/slow camera motion, with limited occlusion.”
BackgroundInfluentialCo-Evolution of Pose and Mesh for 3D Human Body Estimation from Video
You, Y., Liu, H., Wang, T., Li, W., Ding, R., & Li, X. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01374 · Cited by 34
“6M [ 12], and MPI-INF-3DHP [29] contain the annotations of 3D joints and SMPL parameters.”
MethodRecognizing Actions in 3D Using Action-Snippets and Activated Simplices
Wang, C., Flynn, J., Wang, Y., & Yuille, A. (2016) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v30i1.10456 · Cited by 34
“6M (Ionescu et al. 2014).”
Generalizing Monocular 3D Human Pose Estimation in the Wild
Wang, L., Chen, Y., Guo, Z., Qian, K., Lin, M., Li, H., & Ren, J. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00497 · Cited by 34
“6M [12], and HumanEva [34] are collected in indoor scenes and cannot cover various activities.”
BackgroundMethodInfluential3D Human Pose, Shape and Texture from Low-Resolution Images and Videos
Xu, X., Chen, H., Moreno-Noguer, F., Jeni, L. A., & la Torre, F. D. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3070002 · Cited by 34
KAMA: 3D Keypoint Aware Body Mesh Articulation
Iqbal, U., Xie, K., Guo, Y., Kautz, J., & Molchanov, P. A. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00078 · Cited by 35
Liu, R., Shen, J., Wang, H., Chen, C., Cheung, S.-C. S., & Asari, V. K. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01436-0 · Cited by 37
Single image based 3D human pose estimation via uncertainty learning
Han, C., Yu, X., Gao, C., Sang, N., & Yang, Y. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108934 · Cited by 37
Watch It Move: Unsupervised Discovery of 3D Joints for Re-Posing of Articulated Objects
Noguchi, A., Iqbal, U., Tremblay, J., Harada, T., & Gallo, O. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00366 · Cited by 34
“6M dataset [15] do not provide sufficient coverage for our method.”
BackgroundMethodDeltaCNN: End-to-End CNN Inference of Sparse Frame Differences in Videos
Parger, M., Tang, C., Twigg, C. D., Keskin, C., Wang, R. Y., & Steinberger, M. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01217 · Cited by 37
“6M [15] with an input resolution of 384x384.”
BackgroundAdaptPose: Cross-Dataset Adaptation for 3D Human Pose Estimation by Learnable Motion Generation
Gholami, M., Wandt, B., Rhodin, H., Ward, R. K., & Wang, Z. J. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01273 · Cited by 34
“6M [15] dataset on the left and images from the Ski-Pose [33, 35] dataset on the right which we define as source domain and target domain, respectively.”
MethodSource-free Domain Adaptive Human Pose Estimation
Peng, Q., Zheng, C., & Chen, C. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00445 · Cited by 34
“6M [11], one of the target human pose datasets, is a frequently used real-world dataset in the community, consisting of 3.6 million images distributed into seven folds.”
BackgroundInfluentialMVOR: A Multi-view RGB-D Operating Room Dataset for 2D and 3D Human Pose EstimationPreprint
Srivastav, V., Issenhuth, T., Kadkhodamohammadi, A., de Mathelin, M., Gangi, A., & Padoy, N. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.08180 · Cited by 34
“6M[6], HumanEva[13], and IXMAS[15], provide both 2D and 3D annotations, but with a single person performing actions in a controlled environment.”
Consensus-Based Optimization for 3D Human Pose Estimation in Camera Coordinates
Luvizon, D. C., Picard, D., & Tabia, H. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-021-01570-9 · Cited by 33
Learning Perceived Emotion Using Affective and Deep Features for Mental Health Applications
Randhavane, T., Bhattacharya, U., Kapsaskis, K., Gray, K. J., Bera, A., & Manocha, D. (2019) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct.2019.000-2 · Cited by 34
“6M [14]: This dataset is acquired by recording the performance of five female and six male subjects and contains motion-captured gaits from 14 videos of the subjects walking.”
Liu, K., Wan, D., Wang, W., Fei, C., Zhou, T. Y., Guo, D., Bai, L. L., Li, Y., Ni, Z., & Lü, J. (2022) · Advanced Materials · doi:10.1002/adma.202206638 · Cited by 33
Industrial Ergonomics Risk Analysis Based on 3D-Human Pose Estimation
Paudel, P., Kwon, Y.-J., Kim, D., & Choi, K.-H. (2022) · Electronics · doi:10.3390/electronics11203403 · Cited by 34
“It still remains a challenge as some methods use camera array systems to track accurate 3D body motion [35,36] due to occlusion and unclear tracking.”
Monocular 3D Human Pose Estimation for Sports Broadcasts using Partial Sports Field Registration
Baumgartner, T., & Klatt, S. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00539 · Cited by 37
“6m [9].”
BackgroundInfluentialRobots That Can See: Leveraging Human Pose for Trajectory Prediction
Salzmann, T., Chiang, H.-T. L., Ryll, M., Sadigh, D., Parada, C., & Bewley, A. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3312035 · Cited by 33
“M [44], AMASS [45], and 3DPW [46] which are collected using a motion capture system or wearable IMU devices [46], do offer such labels.”
BackgroundHybrid Directed Hypergraph Learning and Forecasting of Skeleton-Based Human Poses
Cui, Q., Ding, Z., & Chen, F. (2024) · Cyborg and Bionic Systems · doi:10.34133/cbsystems.0093 · Cited by 40
“6M [37] includes 15 motion categories, specifically including walking, eating, smoking, discussion, direction, greeting, phoning, posing, purchases, sitting, sitting down, taking photo, waiting, walking dog, and walking together, performed by 7 actors.”
MethodAnimal3D: A Comprehensive Dataset of 3D Animal Pose and Shape
Xu, J., Zhang, Y., Peng, J., Ma, W., Jesslen, A., Ji, P., Hu, Q., Zhang, J., Liu, Q., Wang, J., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00835 · Cited by 43
“6M [13], MPI-INF-3DHP [28], and COCO [25] datasets, and we use the publicly available models to fine-tune them on Animal3D.”
BackgroundMethodYang, Z., Song, D., Ning, J., & Wu, Z. (2024) · IEEE Access · doi:10.1109/access.2024.3509447 · Cited by 33
Survey on Hand Gesture Recognition From Visual Input
Linardakis, M., Varlamis, I., & Papadopoulos, G. T. (2025) · IEEE Access · doi:10.1109/access.2025.3593428 · Cited by 52
CameraHMR: Aligning People with Perspective
Patel, P. D., & Black, M. J. (2025) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv66043.2025.00146 · Cited by 47
“6M [17], COCO [32], MPII [3] and MPI-INF-3DHP [34] while 4DHumans comprises InstaVariety [22], COCO [32], MPII [3], AI Challenger [43] and AVA [14].”
Prediction of Volleyball Trajectory Using Skeletal Motions of Setter Player
Shuya, S., Makino, Y., & Shinoda, H. (2019) · Augmented Human International Conference · doi:10.1145/3311823.3311844 · Cited by 33
“6M human motion dataset [6], which comprises human skeletal data captured using a motion capture system, many types”
MethodHDNet: Human Depth Estimation for Multi-person Camera-Space Localization
Lin, J., & Lee, G. H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58523-5_37 · Cited by 33
Sock, J., Garcia-Hernando, G., Armagan, A., & Kim, T. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00039 · Cited by 37
“Annotating geometric labels such as 6D object pose is costly, often involving special hardware [45, 13], and only captures limited views as the space of possible viewpoints is immense.”
BackgroundVision-Based Gesture Recognition in Human-Robot Teams Using Synthetic Data
de Melo, C. M., Rothrock, B., Gurram, P. K., Ulutan, O., & Manjunath, B. S. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9340728 · Cited by 33
“Synthetic virtual characters have also been used to train depth-based models for static pose estimation [42] [43] [44] [45], gaze estimation [46], and optical flow estimation [47].”
BackgroundMethodAn Empirical Study of the Collapsing Problem in Semi-Supervised 2D Human Pose Estimation
Xie, R., Wang, C., Zeng, W., & Wang, Y. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01105 · Cited by 33
“We extensively evaluate them on three public datasets of COCO [20], MPII [1] and H36M [12].”
MethodInfluentialSkeleton Graph Scattering Networks for 3D Skeleton-based Human Motion Prediction
Li, M., Chen, S.-H., Liu, Z., Zhang, Z., Xie, L.-X., Tian, Q., & Zhang, Y. (2021) · 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw54120.2021.00101 · Cited by 35
“6M dataset [23] has 7 subjects performing 15 classes of actions.”
BackgroundAdversarial Parametric Pose Prior
Davydov, A. V., Remizova, A., Constantin, V., Honari, S., Salzmann, M., & Fua, P. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01072 · Cited by 37
“6M [1] validation subset, chosen according to Protocol 2 (corresponding to the results in Table 4 of the main paper).”
MethodA Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Pose
Zheng, C., Mendieta, M., Wang, P., Lu, A., & Chen, C. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3547844 · Cited by 33
Trajectory Optimization for Physics-Based Reconstruction of 3d Human Pose from Monocular Video
Gärtner, E., Andriluka, M., Xu, H., & Sminchisescu, C. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01276 · Cited by 33
“6M [13], HumanEva-I [32] and a subset of AIST [18, 36].”
BackgroundInfluentialSelf-citationNeuralDome: A Neural Modeling Pipeline on Multi-View Human-Object Interactions
Zhang, J., Luo, H., Yang, H., Xu, X., Wu, Q., Shi, Y., Jingyi Yu, A., Xu, L., & Wang, J. (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/cvpr52729.2023.00853 · Cited by 44
“Early datasets combine multi-camera RGB video capture with synchronized ground-truth 3D skeletons [20, 22, 57], parametric model [25, 67], scanned mesh [15, 78], without human-object interaction.”
BackgroundGaitRef: Gait Recognition with Refined Sequential Skeletons
Zhu, H., Zheng, W., Zheng, Z., & Nevatia, R. (2023) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb57857.2023.10448634 · Cited by 38
“For skeleton refinement, we compare GaitRef with neighbor smoothing (average and gaussian window for neighbor three frames) and SmoothNet [39] (pretrained on H36m [9]) on the CASIA-B dataset with Top-1 accuracy.”
MethodIntention‐aware robot motion planning for safe worker–robot collaboration
Liu, Y., & Jebelli, H. (2023) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.13129 · Cited by 32
“…data set 1 to train and validate the 3D-MotNet, the authors performed an additional step to pretrain the network using two publicly available data sets—the Microsoft Common Objects in Context (COCO) data set (Lin et al., 2014) was used to pretrain the subnetwork 1 of the 3D-MotNet; and the Human3.”
Influential3D skeleton aware driver behavior recognition framework for autonomous driving system
Huo, R., Chen, J., Zhang, Y., & Gao, Q. H. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128743 · Cited by 33
TRAM: Global Trajectory and Motion of 3D Humans from in-the-Wild Videos
Wang, Y., Wang, Z., Liu, L., & Daniilidis, K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73247-8_27 · Cited by 71
Structure from Recurrent Motion: From Rigidity to Recurrency
Li, X., Li, H., Joo, H., Liu, Y., & Sheikh, Y. A. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00320 · Cited by 32
Geissinger, J., & Asbeck, A. T. (2020) · Sensors · doi:10.3390/s20216330 · Cited by 34
Towards Part-Aware Monocular 3D Human Pose Estimation: An Architecture Search Approach
Chen, Z., Huang, Y., Yu, H., Xue, B., Han, K., Guo, Y., & Wang, L. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58580-8_42 · Cited by 33
“6M Dataset [18]: It is captured in a calibrated multi-view studio and consists of 3.6 millions of video frames.”
InfluentialSelf-supervised 3D human pose estimation from video
Gholami, M., Rezaei, A., Rhodin, H., Ward, R. K., & Wang, Z. J. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.02.076 · Cited by 32
Hung-Cuong, N., Nguyen, T.-H., Scherer, R., & Le, V.-H. (2022) · Sensors · doi:10.3390/s22145419 · Cited by 32
MetaPose: Fast 3D Pose from Multiple Views without 3D Supervision
Usman, B., Tagliasacchi, A., Saenko, K., & Sud, A. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00664 · Cited by 33
“Supervised methods [11, 26, 60] yield the lowest 3D pose estimation errors on multi-view single person [24] and multi-person [6, 11, 29] datasets, but require precise camera calibration during both training and inference.”
MethodImplicit Neural Representations With Structured Latent Codes for Human Body Modeling
Peng, S., Chen, G., Zhang, Y., Xu, Y., Wang, Q., Shuai, Q., Zhou, X., & Bao, H. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3245815 · Cited by 32
“6M [18] records multi-view videos with 4 cameras and collects human poses using the marker-based motion capture system.”
MethodInfluentialДу, С., Yuan, Z., Lai, P., & Ikenaga, T. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.110116 · Cited by 32
HybrIK-X: Hybrid Analytical-Neural Inverse Kinematics for Whole-Body Mesh Recovery
Li, J., Bian, S., Xu, C., Chen, Z., Yang, L., & Lu, C. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3528979 · Cited by 68
“6M [17], 3DPW [16], MPI-INF-3DHP [18], and MSCOCO [116] datasets.”
InfluentialPoSynDA: Multi-Hypothesis Pose Synthesis Domain Adaptation for Robust 3D Human Pose Estimation
Liu, H., He, J.-Y., Cheng, Z.-Q., Xiang, W., Yang, Q., Chai, W., Wang, G., Bao, X., Luo, B., Geng, Y., et al. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3612368 · Cited by 31
“6M) [29], MPI-INF-3DHP (3DHP) [46], and 3DPW [56].”
BackgroundVariational autoencoders for 3D data processing
Molnár, S., & Tamás, L. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-023-10687-x · Cited by 36
GCNext: Towards the Unity of Graph Convolutions for Human Motion Prediction
Wang, X., Cui, Q., Chen, C., & Liu, M. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28375 · Cited by 31
A Kinematic Chain Space for Monocular Motion Capture
Wandt, B., Ackermann, H., & Rosenhahn, B. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11018-5_4 · Cited by 31
3D human pose estimation with siamese equivariant embedding
Véges, M., Varga, V., & Lörincz, A. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.02.029 · Cited by 33
MetaFuse: A Pre-trained Fusion Model for Human Pose Estimation
Xie, R., Wang, C., & Wang, Y. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.01370 · Cited by 31
Metadata
- Authors
- Ionescu, C., Papava, D., Olaru, V., & Sminchisescu, C.
- Affiliations
- IMAR - Institutul de Matematică „Simion Stoilow” al Academiei Române (RO)
- Lund University (SE)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2014-07
- Volume
- 36
- Issue
- 7
- Pages
- 1325-1339
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- OpenAlex
- W2101032778
- Cited by
- 3,653
- References
- 71
- Citations by year
- 2026: 349 · 2025: 529 · 2024: 581 · 2023: 543 · 2022: 390 · 2021: 429 · 2020: 305 · 2019: 235 · 2018: 133 · 2017: 78 · 2016: 48 · 2015: 20 · 2014: 10
- FWCI
- 46.29
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- Lund University Publications (Lund University) (submittedVersion)
- Funders
- Universitatea Politehnica din București
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Physical Sciences › Computer Science › Human-Computer Interaction › Hand Gesture Recognition Systems
- Keywords
- 3D human pose estimation, natural environment, 3D body scans, occlusion, human activities
- MeSH terms
- Algorithms, Biometry, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Pattern Recognition, Automated, Photography, Posture, Sensitivity and Specificity, Subtraction Technique, Reproducibility of Results, Databases, Factual, Information Storage and Retrieval, Ecosystem, Imaging, Three-Dimensional, Whole Body Imaging
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
