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 301–400 of 1,009
Human Motion Generation: A Survey
Zhu, W., Ma, X., Ro, D., Ci, H., Zhang, J., Shi, J., Gao, F., Tian, Q. C., & Wang, Y. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3330935 · Cited by 141
“On the other hand, the remarkable progress in human modeling [33], [34], [35] makes it easier to extract human motion from videos [36], [37], [38] and construct large-scale human motion datasets [39], [40], [41], [42].”
BackgroundMethodMarker-Less 3D Human Motion Capture with Monocular Image Sequence and Height-Maps
Du, Y., Wong, Y., Liu, Y., Han, F., Gui, Y., Wang, Z., Kankanhalli, M., & Geng, W. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46493-0_2 · Cited by 109
“The error are reported in mean per joint position error (MPJPE) [45].”
BackgroundMethodInfluentialLearning disentangled representations in the imaging domain
Liu, X., Sanchez, P., Thermos, S., O’Neil, A. Q., & Tsaftaris, S. A. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102516 · Cited by 130
“6M (Ionescu et al., 2014) and Penn Action (Zhang et al., 2013) human activity datasets datasets.”
CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wild
Wandt, B., Rudolph, M., Zell, P., Rhodin, H., & Rosenhahn, B. (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/cvpr46437.2021.01309 · Cited by 108
“6M [10] and MPI-INF-3DHP [24] and set the new state of the art in several metrics for self-supervised 3D pose estimation.”
BackgroundBeLFusion: Latent Diffusion for Behavior-Driven Human Motion Prediction
Barquero, G., Escalera, S., & Palmero, C. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00220 · Cited by 128
“To summarize, our main contributions are: (1) We pro-pose BeLFusion, a method that generates predictions that are significantly more realistic and coherent with the near past than prior works, while achieving state-of-the-art accuracy on Human 3.6M [32] and AMASS [43] datasets.”
WHAM: Reconstructing World-Grounded Humans with Accurate 3D Motion
Shin, S., Kim, J.-Y., Halilaj, E., & Black, M. J. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00202 · Cited by 164
“We use a pretrained image encoder from previous work [1, 6, 20, 24] and train the feature integrator using video datasets [10, 14, 32, 47].”
MethodWho Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop
Biggs, B., Boyne, O., Charles, J., Fitzgibbon, A., & Cipolla, R. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58621-8_12 · Cited by 107
Li, L., Martin, T., & Xu, X. (2020) · Applied Ergonomics · doi:10.1016/j.apergo.2020.103138 · Cited by 111
DHP19: Dynamic Vision Sensor 3D Human Pose Dataset
Calabrese, E., Taverni, G., Easthope, C. A., Skriabine, S., Corradi, F., Longinotti, L., Eng, K., & Delbruck, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00217 · Cited by 99
“6M [16] F 4 11 15 MPJPE1 MPI-INF-3DHP [22] F 14 8 8 MPJPE1, PCK2 MADS [32] F 3 5 30 MPJPE1 DHP19 (This work) E 4 17 33 MPJPE1”
Multiple human 3D pose estimation from multiview images
Ershadi-Nasab, S., Noury, E., Shohreh Kasaei, B., & Sanaei, E. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-5133-8 · Cited by 105
“6M [29], KTH Football II [15], and MPII Cooking [44] datasets.”
BackgroundMethod3D Human Sensing, Action and Emotion Recognition in Robot Assisted Therapy of Children with Autism
Marinoiu, E., Zanfir, M., Olaru, V., & Sminchisescu, C. (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.2018.00230 · Cited by 106
“DMHS is trained on fully visible humans from Human80K [14], a subset of [15], which contains data for 11 adult actors performing 15 different actions in a laboratory setup.”
BackgroundSelf-citationHuman Motion Prediction via Learning Local Structure Representations and Temporal Dependencies
Guo, X. J., & Choi, J. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33012580 · Cited by 106
Recurrent 3D Pose Sequence Machines
Lin, M., Lin, L., Liang, X., Wang, K., & Cheng, H. (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.2017.588 · Cited by 110
Zanfir, M., Popa, A.-I., Zanfir, A., & Sminchisescu, C. (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.2018.00565 · Cited by 105
Self-citationNeural monocular 3D human motion capture with physical awareness
Shimada, S., Golyanik, V., Xu, W., Pérez, P., & Theobalt, C. (2021) · ACM Transactions on Graphics · doi:10.1145/3450626.3459825 · Cited by 108
Can 3D Pose Be Learned from 2D Projections Alone?
Drover, D., Rohith, M. V., Chen, C.-H., Agrawal, A., Tyagi, A., & Huynh, C. P. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11018-5_7 · Cited by 107
AIFit: Automatic 3D Human-Interpretable Feedback Models for Fitness Training
Fieraru, M., Zanfir, M., Pirlea, S. C., Olaru, V., & Sminchisescu, C. (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/cvpr46437.2021.00979 · Cited by 112
“6M [10], a large-scale 3d dataset capturing everyday activities.”
BackgroundSelf-citationSmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos
Zeng, A., Yang, L., Ju, X., Li, J., Wang, J., & Xu, Q. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20065-6_36 · Cited by 115
HumanMAC: Masked Motion Completion for Human Motion Prediction
Chen, L.-H., Zhang, J., Li, Y., Pang, Y., Xia, X., & Liu, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00875 · Cited by 119
“6M [27] and HumanEva-I [62].”
Zhang, W., Liu, Z., Zhou, L., Leung, H. W. H., & Chan, A. B. (2017) · Image and Vision Computing · doi:10.1016/j.imavis.2017.02.002 · Cited by 114
I Know That Person: Generative Full Body and Face De-identification of People in Images
Brkić, K., Sikiric, I., Hrkać, T., & Kalafatić, Z. (2017) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.173 · Cited by 103
“6M dataset [13].”
MeTRAbs: Metric-Scale Truncation-Robust Heatmaps for Absolute 3D Human Pose Estimation
Sarandi, I., Linder, T., Arras, K. O., & Leibe, B. (2020) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2020.3037257 · Cited by 106
ARAH: Animatable Volume Rendering of Articulated Human SDFs
Wang, S., Schwarz, K., Geiger, A., & Tang, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19824-3_1 · Cited by 112
Three-Dimensional Reconstruction of Human Interactions
Fieraru, M., Zanfir, M., Oneaţă, E., Popa, A.-I., Olaru, V., & Sminchisescu, C. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00724 · Cited by 114
“Most datasets dedicated to human understanding are centered around single person scenarios[1, 13, 38, 25] and even those that include multiple people [20, 46, 7] do not explicitly model the close interaction between different people.”
BackgroundSelf-citationIterated Second-Order Label Sensitive Pooling for 3D Human Pose Estimation
Ionescu, C., Carreira, J., & Sminchisescu, C. (2014) · 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.2014.215 · Cited by 101
“the arm consists of the shoulder, elbow, wrist as well as humerus and radius), 2 for the torso (chest and abdomen), 1 for the pelvis and 1 for the head (see [18] for details).”
BackgroundMethodInfluentialSelf-citationImitation Learning for Human Pose Prediction
Wang, B., Adeli, E., Chiu, H.-K., Huang, D.-A., & Niebles, J. C. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00722 · Cited by 104
“6M dataset [19] using: (a) BC alone; (b) WGAIL-div alone; and (c) WGAIL-div with BC as pretraining.”
3D Human Pose Estimation via Intuitive Physics
Tripathi, S., Müller, L., Huang, C.-H. P., Taheri, O., Black, M. J., & Tzionas, D. (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/cvpr52729.2023.00457 · Cited by 116
GraphMLP: A graph MLP-like architecture for 3D human pose estimation
Li, W., Liu, M., Liu, H., Guo, T., Wang, T., Tang, H., & Sebe, N. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110925 · Cited by 116
4D Association Graph for Realtime Multi-Person Motion Capture Using Multiple Video Cameras
Zhang, Y., An, L., Yu, T., Li, X., Li, K., & Liu, 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/cvpr42600.2020.00140 · Cited by 100
3D Human Pose Estimation With 2D Marginal Heatmaps
Nibali, A., He, Z., Morgan, S., & Prendergast, L. A. (2019) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2019.00162 · Cited by 98
Human Body Model Fitting by Learned Gradient Descent
Song, J., Chen, X., & Hilliges, O. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58565-5_44 · Cited by 100
Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded Scenes
Choi, H., Moon, G., Park, J., & Lee, K. 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/cvpr52688.2022.00153 · Cited by 111
“6M dataset [6]’s licenses are limited to academic use only.”
BackgroundResultInfluentialH-NeRF: Neural Radiance Fields for Rendering and Temporal Reconstruction of Humans in MotionPreprint
Xu, H., Alldieck, T., & Sminchisescu, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.13746 · Cited by 101
“6M [20] (5 sequences).”
Self-citation3D Human Pose Machines with Self-supervised Learning
Wang, K., Lin, L., Jiang, C., Qian, C., & Wei, P. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2892452 · Cited by 104
Putting Humans in a Scene: Learning Affordance in 3D Indoor Environments
Li, X., Liu, S., Kim, K., Wang, X., Yang, M., & Kautz, J. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.01265 · Cited by 96
3D Human Mesh Regression With Dense Correspondence
Zeng, W., Ouyang, W., Luo, P., Liu, W., & Wang, X. (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/cvpr42600.2020.00708 · Cited by 101
A Comprehensive Survey on Deep Learning Methods in Human Activity Recognition
Kaseris, M., Kostavelis, I., & Malassiotis, S. (2024) · Machine Learning and Knowledge Extraction · doi:10.3390/make6020040 · Cited by 106
“A common way to represent a skeleton is illustrated in Figure 8 and typically is represented in the form of joint rotations in either exponential maps or quaternions [41,42].”
BackgroundimGHUM: Implicit Generative Models of 3D Human Shape and Articulated Pose
Alldieck, T., Xu, H., & Sminchisescu, C. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00541 · Cited by 102
“6M [20] and the CMU mocap dataset [2], including head and hand poses.”
BackgroundSelf-citationGenerating Multiple Diverse Hypotheses for Human 3D Pose Consistent with 2D Joint Detections
Jahangiri, E., & Yuille, A. (2017) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2017.100 · Cited by 97
“6M dataset [15].”
BackgroundVision-Based Human Pose Estimation via Deep Learning: A Survey
Lan, G., Wu, Y., Hu, F., & Hao, Q. (2022) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2022.3219242 · Cited by 109
“proaches since the existing datasets 3-D HPE is mostly collected by indoor motion capture systems [33], [34].”
BackgroundVision-Based Estimation of MDS-UPDRS Gait Scores for Assessing Parkinson’s Disease Motor Severity
Lu, M., Poston, K. L., Pfefferbaum, A., Sullivan, E. V., Fei-Fei, L., Pohl, K. M., Niebles, J. C., & Adeli, E. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-59716-0_61 · Cited by 95
HuMMan: Multi-modal 4D Human Dataset for Versatile Sensing and Modeling
Cai, Z., Ren, D., Zeng, A., Lin, Z., Yu, T., Wang, W., Fan, X., Gao, Y., Yu, Y., Pan, L., et al. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20071-7_33 · Cited by 110
Human Pose as Compositional Tokens
Geng, Z., Wang, C., Wei, Y., Liu, Z., Li, H., & Hu, H. (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/cvpr52729.2023.00071 · Cited by 104
“6M [32] dataset which has 11 human subjects performing daily actions.”
BackgroundMethodInfluentialLiu, H., Zhu, Z., Becherini, G., Peng, Y., Su, M., Zhou, Y., Zhe, X., Iwamoto, N., Zheng, B., & Black, M. J. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00115 · Cited by 145
“For instance, the average error (in mm) for the SOTA monocular 3D pose estimation algorithm [23] in the Human 3.6M dataset [31] is 33.4, while for Vicon mocap it is 0.142 [47].”
BackgroundInfluentialCompressed Volumetric Heatmaps for Multi-Person 3D Pose Estimation
Fabbri, M., Lanzi, F., Calderara, S., Alletto, S., & Cucchiara, R. (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/cvpr42600.2020.00723 · Cited by 97
“This allows our method to achieve state of the art performance on the most popular single person benchmark [11].”
MethodDeep network for the integrated 3D sensing of multiple people in natural images
Zanfir, A., Marinoiu, E., Zanfir, M., Popa, A. I., & Sminchisescu, C. (2018) · Lund University Publications (Lund University) · Cited by 93
Self-citationTowards Social Artificial Intelligence: Nonverbal Social Signal Prediction in a Triadic Interaction
Joo, H., Simon, T., Cikara, M., & Sheikh, Y. A. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.01113 · Cited by 100
“Datasets providing rich 3D body motion information captured with motion capture techniques exist, but they contain single subjects’ motion only [24, 30, 62].”
BackgroundPredicting 3D Human Dynamics From Video
Zhang, J. Y., Felsen, P., Kanazawa, A., & Malik, J. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00721 · Cited by 93
“6M [23], which contains videos of actors performing various activities in a motion capture studio.”
BackgroundEfficient human motion prediction using temporal convolutional generative adversarial network
Cui, Q., Sun, H., Kong, Y., Zhang, X., & Li, Y. (2020) · Information Sciences · doi:10.1016/j.ins.2020.08.123 · Cited by 96
Encoder-decoder with Multi-level Attention for 3D Human Shape and Pose Estimation
Wan, Z., Li, Z., Tian, M., Liu, J., Yi, S., & Li, H. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01279 · Cited by 96
Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation
Moon, G., Choi, H., & Lee, K. M. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00257 · Cited by 119
DropKey for Vision Transformer
Li, B., Hu, Y., Nie, X., Han, C., Jiang, X., Guo, T., & Liu, L. (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/cvpr52729.2023.02174 · Cited by 106
“We use Mean Per Vetex Error (MPVE) [22], Mean Per Joint Position Error (MPJPE) [14] and Procrustes Analysis MPJPE (PA-MPJPE) [38] as metrics.”
MethodDiverse Trajectory Forecasting with Determinantal Point ProcessesPreprint
Yuan, Y., & Kitani, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.04967 · Cited by 93
“6M (Ionescu et al., 2013), to evaluate the generalization ability of our approach.”
BackgroundChallenges in Multi-modal Gesture Recognition
Escalera, S., Athitsos, V., & Guyon, I. (2017) · The Springer series on challenges in machine learning · doi:10.1007/978-3-319-57021-1_1 · Cited by 91
“In (Ionescu et al., 2014), it is presented the Human3.”
Hierarchical Kinematic Human Mesh Recovery
Georgakis, G., Li, R., Karanam, S., Chen, T., Košecká, J., & Wu, Z. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58520-4_45 · Cited by 104
Autonomous aerial cinematography in unstructured environments with learned artistic decision‐making
Bonatti, R., Wang, W., Ho, C., Ahuja, A., Gschwindt, M., Camci, E., Kayacan, E., Choudhury, S., & Scherer, S. (2020) · Journal of Field Robotics · doi:10.1002/rob.21931 · Cited by 102
Posture-related data collection methods for construction workers: A review
Yu, Y., Umer, W., Yang, X., & Antwi‐Afari, M. F. (2021) · Automation in Construction · doi:10.1016/j.autcon.2020.103538 · Cited by 91
Bhattacharya, U., Childs, E., Rewkowski, N., & Manocha, D. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475223 · Cited by 91
“6M dataset [60] of human motions using the ten joints in the TED Gesture Dataset [1].”
MethodWhen XR and AI Meet - A Scoping Review on Extended Reality and Artificial Intelligence
Hirzle, T., Müller, F., Draxler, F., Schmitz, M., Knierim, P., & Hornbæk, K. (2023) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3581072 · Cited by 101
Imposing Hard Constraints on Deep Networks: Promises and LimitationsPreprint
Márquez-Neila, P., Salzmann, M., & Fua, P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.02025 · Cited by 91
Multi-level Motion Attention for Human Motion Prediction
Mao, W., Liu, M., Salzmann, M., & Li, H. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01483-7 · Cited by 90
VoxelTrack: Multi-Person 3D Human Pose Estimation and Tracking in the Wild
Zhang, Y., Wang, C., Wang, X., Liu, W., & Zeng, W. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3163709 · Cited by 91
Pose Transformers (POTR): Human Motion Prediction with Non-Autoregressive Transformers
Martinez-Gonzalez, A. N., Villamizar, M., & Odobez, J. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00257 · Cited by 90
Investigating Pose Representations and Motion Contexts Modeling for 3D Motion Prediction
Liu, Z., Wu, S., Jin, S., Ji, S., Liu, Q., Lu, S., & Cheng, L. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3139918 · Cited by 91
Adaptive Multi-View and Temporal Fusing Transformer for 3D Human Pose Estimation
Shuai, H., Wu, L., & Liu, Q. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3188716 · Cited by 90
“6M [29], TotalCapture [30], and KTH Multiview Football II [31] quantitatively and qualitatively.”
BackgroundInfluentialRethinking Pose in 3D: Multi-stage Refinement and Recovery for Markerless Motion Capture
Tomè, D., Toso, M., Agapito, L., & Russell, C. (2018) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2018.00061 · Cited by 89
Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Images
Liang, J., & Lin, M. C. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00445 · Cited by 89
“6M [17] datasets.”
Quantifying Parkinson’s disease motor severity under uncertainty using MDS-UPDRS videos
Lu, M., Zhao, Q., Poston, K. L., Sullivan, E. V., Adolf Pfefferbaum, A., Shahid, M., Katz, M., Montaser‐Kouhsari, L., Schulman, K. A., Milstein, A. S., et al. (2021) · Medical Image Analysis · doi:10.1016/j.media.2021.102179 · Cited by 97
“6M (Ionescu et al., 2013) and MPI-INF-3DHP (Mehta et al., 2017) datasets, providing over 150k training images with 3D joint annotations, as well as large-scale datasets with 2D annotations ( e.g ., COCO (Lin et al., 2017) and MPII (Andriluka et al., 2014)).”
MethodSimultaneously-Collected Multimodal Lying Pose Dataset: Enabling In-Bed Human Pose Monitoring
Liu, S., Huang, X., Fu, N., Li, C., Su, Z., & Ostadabbas, S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3155712 · Cited by 103
“As far as 3D human pose estimation, noticeable improvements have also been achieved either by the end-to-end training on real 3D human datasets [21] or based on a learned human body template [22].”
BackgroundEfficiently Creating 3D Training Data for Fine Hand Pose Estimation
Oberweger, M., Riegler, G., Wohlhart, P., & Lepetit, V. (2016) · 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.2016.536 · Cited by 88
“od, leading to increased accuracy. The code and dataset can be found on our website https://cvarlab.icg.tugraz. at/projects/hand_detection/. 1. Introduction Recent work on articulated pose estimation [7,16,24, 26,28] has shown that a large amount of accurate training data makes reliable and precise estimation possible…”
BackgroundLearning Rigidity in Dynamic Scenes with a Moving Camera for 3D Motion Field Estimation
Lv, Z., Kim, K., Troccoli, A. J., Sun, D., Rehg, J. M., & Kautz, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01228-1_29 · Cited by 88
HUMBI: A Large Multiview Dataset of Human Body Expressions
Yu, Z., Yoon, J. S., Lee, I. K., Venkatesh, P., Park, J., Yu, J., & Park, H. S. (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/cvpr42600.2020.00306 · Cited by 96
MeTRAbs: Metric-Scale Truncation-Robust Heatmaps for Absolute 3D Human Pose Estimation
Sárándi, I., Linder, T., Arras, K. O., & Leibe, B. (2021) · RWTH Publications (RWTH Aachen) · Cited by 88
Physics-based Human Motion Estimation and Synthesis from Videos
Xie, K., Wang, T., Iqbal, U., Guo, Y., Fidler, S., & Shkurti, F. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01133 · Cited by 88
“6M [12], 3DPW [49] and MSCOCO [25] datasets.”
MethodInfluentialGenerating Smooth Pose Sequences for Diverse Human Motion Prediction
Mao, W., Liu, M., & Salzmann, M. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01306 · Cited by 102
“6M [26] and HumanEva-I [47], and adopt the same training and testing settings as in [55] on both datasets.”
MethodBiderman, D., Whiteway, M. R., Hurwitz, C., Greenspan, N., Lee, R. S., Vishnubhotla, A., Warren, R., Pedraja, F., Noone, D., Schartner, M. M., et al. (2024) · Nature Methods · doi:10.1038/s41592-024-02319-1 · Cited by 104
“Human models are trained on much larger labeled datasets containing either annotated images 49 or 3D motion capture 50 .”
Graph Convolutional Networks for Assessment of Physical Rehabilitation Exercises
Deb, S., Islam, M. F., Rahman, S., & Rahman, S. (2022) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2022.3150392 · Cited by 88
EgoBody: Human Body Shape and Motion of Interacting People from Head-Mounted Devices
Zhang, S., Ma, Q., Zhang, Y., Qian, Z., Kwon, T., Pollefeys, M., Bogo, F., & Tang, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20068-7_11 · Cited by 104
“A large number of datasets focus on 3D human pose and motion from third-person-views [23,30,34, 37,38,52,63–65,72,86,97,103].”
BackgroundMethodInfluentialCOUCH: Towards Controllable Human-Chair Interactions
Zhang, X., Bhatnagar, B. L., Starke, S., Guzov, V., & Pons‐Moll, G. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20065-6_30 · Cited by 87
A generalizable approach for multi-view 3D human pose regression
Kadkhodamohammadi, A., & Padoy, N. (2020) · Machine Vision and Applications · doi:10.1007/s00138-020-01120-2 · Cited by 88
“uman3.6M and MS COCO that is used to train the detector. The detector is applied on the whole image, i.e. no bounding box is provided, in contrast to previous work that relies either on ground truth (Ionescu et al., 2014; Moreno-Noguer, 2017; Martinez et al., 2017) or on person detectors (Tekin et al., 2016b) to obtai…”
BackgroundMethodInfluentialEstimating Parkinsonism Severity in Natural Gait Videos of Older Adults With Dementia
Sabo, A., Mehdizadeh, S., Iaboni, A., & Taati, B. (2021) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2022.3144917 · Cited by 67
AI-Driven Stroke Rehabilitation Systems and Assessment: A Systematic Review
Rahman, S., Sarker, S. C., Haque, A. K. M. N., Uttsha, M. M., Islam, M. F., & Deb, S. (2022) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2022.3219085 · Cited by 99
“6M [121]).”
Khirodkar, R., Tripathi, S., & Kitani, K. (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/cvpr52688.2022.00176 · Cited by 88
“Most benchmark datasets [18, 21, 23, 46, 55, 61, 63] used for learning human mesh recovery focus on a single person and do not accurately represent the distribution of possible occlusions present in the real world.”
BackgroundCamera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D Registration
Chen, X., Liu, Y., Ma, C., Chang, J., Wang, H., Chen, T., Guo, X., Wan, P., & Zheng, W. (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/cvpr46437.2021.01307 · Cited by 90
“6M [14] is a large-scale 3D body pose benchmark containing 3.”
BackgroundSkeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction
Li, M., Chen, S., Zhang, Z., Xie, L., Tian, Q., & Zhang, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20068-7_2 · Cited by 87
Perpetual Humanoid Control for Real-time Simulated Avatars
Luo, Z., Cao, J., Winkler, A., Kitani, K., & Xu, W. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01000 · Cited by 140
“To evaluate our policy’s ability to handle unseen MoCap sequences and noisy pose estimate from pose estimation methods, we use the popular H36M dataset [16].”
MethodSocially and Contextually Aware Human Motion and Pose Forecasting
Adeli, V., Adeli, E., Reid, I. D., Niebles, J. C., & Rezatofighi, H. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3010742 · Cited by 85
Greedy Hierarchical Variational Autoencoders for Large-Scale Video Prediction
Wu, B.-H., Nair, S., Martín-Martín, R., Fei-Fei, L., & Finn, C. (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/cvpr46437.2021.00235 · Cited by 90
“6M [17] to assess prediction of structured motion.”
BackgroundMultiscale Spatio-Temporal Graph Neural Networks for 3D Skeleton-Based Motion Prediction
Li, M., Chen, S., Zhao, Y., Zhang, Y., Wang, Y., & Tian, Q. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3108708 · Cited by 86
Decoupling Human and Camera Motion from Videos in the Wild
Ye, V., Pavlakos, G., Malik, J., & Kanazawa, 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/cvpr52729.2023.02033 · Cited by 100
Dynamic Dense Graph Convolutional Network for Skeleton-Based Human Motion Prediction
Wang, X., Zhang, W., Wang, C., Gao, Y., & Liu, M. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3334954 · Cited by 90
“6M) [70] is a large-scale dataset which involves 15 types of actions performed by 7 actors.”
BackgroundMethodDiffPose: Multi-hypothesis Human Pose Estimation using Diffusion Models
Holmquist, K., & Wandt, B. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01464 · Cited by 99
“6M [15] and MPI-INF-3DHP [32] using their established training and test splits.”
MethodHigh Fidelity Video Prediction with Large Stochastic Recurrent Neural NetworksPreprint
Villegas, R., Pathak, A., Kannan, H., Erhan, D., Le, Q. V., & Lee, H. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.01655 · Cited by 89
“We use the Human 3.6M dataset [Ionescu et al., 2014] to measure the ability of our models to predict structured motion.”
Context Modeling in 3D Human Pose Estimation: A Unified Perspective
Ma, X., Su, J., Wang, C., Ci, H., & Wang, Y. (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/cvpr46437.2021.00617 · Cited by 83
“Results on the H36M Dataset Table 2 shows the results of the state-of-the-art methods on the H36M dataset.”
MethodInfluentialHOGFormer: high-order graph convolution transformer for 3D human pose estimation
Xie, Y., Hong, C., Zhuang, W., Liu, L., & Li, J. (2024) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-024-02262-9 · Cited by 83
“6M [38] is one of the most widely used datasets for 3D HPE, containing 3.6 million images.”
InfluentialDepth Sweep Regression Forests for Estimating 3D Human Pose from Images
Kostrikov, I., & Gall, J. (2014) · British Machine Vision Conference (BMVC) · doi:10.5244/c.28.80 · Cited by 83
PPT: Token-Pruned Pose Transformer for Monocular and Multi-view Human Pose Estimation
Ma, H., Wang, Z., Chen, Y., Kong, D., Chen, L., Liu, X., Yan, X., Tang, H., & Xie, X. S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20065-6_25 · Cited by 90
Spatio-Temporal Manifold Learning for Human Motions via Long-Horizon Modeling
Wang, H., Ho, E. S. L., Shum, H. P. H., & Zhu, Z. (2019) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2019.2936810 · Cited by 90
“t be a good metric for motion prediction of a specific type but a good indicator to show how quickly errors accumulate. We follow the protocol in [6] and compare our model with others on H3.6M dataset [49]. We randomly choose 8 prefixes from the dataset and compare our prediction errors with some stateof-the-art methods…”
ResultFusing Wearable IMUs With Multi-View Images for Human Pose Estimation: A Geometric Approach
Zhang, Z., Wang, C., Qin, W., & Zeng, W. (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/cvpr42600.2020.00227 · Cited by 81
Towards Accurate 3D Human Motion Prediction from Incomplete Observations
Cui, Q., & Sun, H. (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/cvpr46437.2021.00477 · Cited by 82
“Then, following the previous work [41, 13, 33], we also provide 3D errors using Mean Per Joint Position Error (MPJPE) [25] in millimeter (mm).”
BackgroundMethod
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
