</> Code with Papers by Syntology

Datasets › Human3.6M

Human3.6M

Introduced by Catalin Ionescu et al. in Human3.6m: Large scale datasets and predictive methods for 3D human sensing in natural environments1 Jan 2014 archive 2025-07-28

The Human3.6M dataset is one of the largest motion capture datasets, which consists of 3.6 million human poses and corresponding images captured by a high-speed motion capture system. There are 4 high-resolution progressive scan cameras to acquire video data at 50 Hz. The dataset contains activities by 11 professional actors in 17 scenarios: discussion, smoking, taking photo, talking on the phone, etc., as well as provides accurate 3D joint positions and high-resolution videos.

Source: Space-Time Representation of People Based on 3D Skeletal Data: A Review

Image Source: Yu et al

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 13 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
3D Human Pose Estimation Human3.6M LMT R152 384x384 Average MPJPE (mm) 17.59 Learnable human mesh triangulation for 3D human pose and... — 88 Compare
Monocular 3D Human Pose Estimation Human3.6M MotionBERT (Finetune) Average MPJPE (mm) 37.5 MotionBERT: A Unified Perspective on Learning Human... Walter0807/MotionBERT 52 Compare
Human Pose Forecasting Human3.6M GCNext Average MPJPE (mm) @ 400ms 30.5 GCNext: Towards the Unity of Graph Convolutions for... bradleywang0416/gcnext 33 Compare
Weakly-supervised 3D Human Pose Estimation Human3.6M AdaptPose Average MPJPE (mm) 42.5 AdaptPose: Cross-Dataset Adaptation for 3D Human Pose... mgholamikn/AdaptPose 33 Compare
Multi-Hypotheses 3D Human Pose Estimation Human3.6M DDHPose (H=20, W=10, J-Best) Average MPJPE (mm) 33.62 Disentangled Diffusion-Based 3D Human Pose Estimation... Andyen512/DDHPose 12 Compare
Unsupervised 3D Human Pose Estimation Human3.6M ElePose MPJPE 64.0 ElePose: Unsupervised 3D Human Pose Estimation by... bastianwandt/elepose 12 Compare
Video Prediction Human3.6M IAM4VP SSIM 0.942 Implicit Stacked Autoregressive Model for Video Prediction seominseok0429/Implicit-Stacked-Autoregressive-Model-for-Video-Prediction 9 Compare
Human action generation Human3.6M Kinetic-GAN MMDa 0.071 Generative Adversarial Graph Convolutional Networks for... degardinbruno/kinetic-gan 5 Compare
3D Absolute Human Pose Estimation Human3.6M GnTCN MRPE 88.1 Graph and Temporal Convolutional Networks for 3D... 3dpose/GnTCN 4 Compare
3D Pose Estimation Human3.6M TesseTrack Average MPJPE (mm) 18.7 TesseTrack: End-to-End Learnable Multi-Person... — 4 Compare
Human Part Segmentation Human3.6M Hulk(Finetune, ViT-L) mIoU 69.89 Hulk: A Universal Knowledge Translator for Human-Centric Tasks opengvlab/humanbench +1 3 Compare
2D Pose Estimation Human3.6M UniHCP (finetune) EPE 6.6 UniHCP: A Unified Model for Human-Centric Perceptions opengvlab/unihcp 1 Compare
Pose Retrieval Human3.6M Pr-VIPE Hit@1 76.2 View-Invariant Probabilistic Embedding for Human Pose google-research/google-research +1 1 Compare

Papers archive 2025-07-28

30 shown of 175 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 783. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation 1 2 3 Jun 2025 not harvested
GenHMR: Generative Human Mesh Recovery 0 1 19 Dec 2024 not harvested
VoxelKeypointFusion: Generalizable Multi-View Multi-Person Pose Estimation 1 1 24 Oct 2024 not harvested
KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimation 1 2 31 Mar 2024 ran 6 of 10 samples (4 unverified; 10 pointer-only for licence)
A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation 1 2 17 Mar 2024 ran 2 of 5 samples (3 unverified; 5 pointer-only for licence)
Disentangled Diffusion-Based 3D Human Pose Estimation with Hierarchical Spatial and Temporal Denoiser 1 3 7 Mar 2024 not harvested
Geometry-Biased Transformer for Robust Multi-View 3D Human Pose Reconstruction 0 1 28 Dec 2023 not harvested
GCNext: Towards the Unity of Graph Convolutions for Human Motion Prediction 1 1 19 Dec 2023 not harvested
SoloPose: One-Shot Kinematic 3D Human Pose Estimation with Video Data Augmentation 1 1 15 Dec 2023 not harvested
Mask as Supervision: Leveraging Unified Mask Information for Unsupervised 3D Pose Estimation 1 1 12 Dec 2023 not harvested
Hulk: A Universal Knowledge Translator for Human-Centric Tasks 2 2 4 Dec 2023 ran 12 of 24 samples (12 unverified)
MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Network 1 4 25 Oct 2023 ran 14 of 16 samples (2 unverified)
LInKs "Lifting Independent Keypoints" -- Partial Pose Lifting for Occlusion Handling with Improved Accuracy in 2D-3D Human Pose Estimation 0 1 13 Sep 2023 not harvested
AnyPose: Anytime 3D Human Pose Forecasting via Neural Ordinary Differential Equations 0 1 9 Sep 2023 not harvested
Multiscale Residual Learning of Graph Convolutional Sequence Chunks for Human Motion Prediction 1 1 31 Aug 2023 not harvested
Auxiliary Tasks Benefit 3D Skeleton-based Human Motion Prediction 1 1 17 Aug 2023 ran 13 of 19 samples (6 unverified; 19 pointer-only for licence)
A Unified Masked Autoencoder with Patchified Skeletons for Motion Synthesis 0 1 14 Aug 2023 not harvested
Double-chain Constraints for 3D Human Pose Estimation in Images and Videos 1 1 10 Aug 2023 not harvested
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction 1 1 30 Jul 2023 not harvested
Fast Fourier Inception Networks for Occluded Video Prediction 1 1 17 Jun 2023 not harvested
Remembering What Is Important: A Factorised Multi-Head Retrieval and Auxiliary Memory Stabilisation Scheme for Human Motion Prediction 0 1 19 May 2023 not harvested
Towards Accurate Human Motion Prediction via Iterative Refinement 0 1 8 May 2023 not harvested
TAPE: Temporal Attention-based Probabilistic human pose and shape Estimation 1 1 29 Apr 2023 not harvested
Sampling is Matter: Point-guided 3D Human Mesh Reconstruction 1 1 19 Apr 2023 ran 5 of 8 samples (3 unverified)
Multi-Graph Convolution Network for Pose Forecasting 0 1 11 Apr 2023 not harvested
Graph-Guided MLP-Mixer for Skeleton-Based Human Motion Prediction 0 1 7 Apr 2023 not harvested
ConvFormer: Parameter Reduction in Transformer Models for 3D Human Pose Estimation by Leveraging Dynamic Multi-Headed Convolutional Attention 1 1 4 Apr 2023 not harvested
Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregation 1 2 21 Mar 2023 ran 0 of 2 samples (2 unverified)
EqMotion: Equivariant Multi-agent Motion Prediction with Invariant Interaction Reasoning 2 1 20 Mar 2023 ran 1 of 2 samples (1 unverified)
Implicit Stacked Autoregressive Model for Video Prediction 1 1 14 Mar 2023 not harvested

The full list of 175 is in the JSON twin.

Dataset loaders archive 2025-07-28

open-mmlab/mmposeopen-mmlab/mmposepytorch
gitlab.com/Percipiote/skelda

2 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

3D Human Pose Estimation3D Absolute Human Pose EstimationVideo Prediction2D Pose EstimationHuman Pose Forecasting3D Pose EstimationHuman action generationHuman Part SegmentationMulti-Hypotheses 3D Human Pose EstimationUnsupervised Human Pose EstimationPose RetrievalWeakly-supervised 3D Human Pose EstimationUnsupervised 3D Human Pose EstimationRoot Joint LocalizationMonocular 3D Human Pose Estimation3D Human Pose Estimation in Limited Data

License archive 2025-07-28

Custom (research-only, attribution)

Modalities archive 2025-07-28

ImagesVideos

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • Human3.6M

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections