Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

computer-vision-1

3D Human Pose Estimation

3260 papers • 126 benchmarks • 313 datasets

3D Human Pose Estimation is a computer vision task that involves estimating the 3D positions and orientations of body joints and bones from 2D images or videos. The goal is to reconstruct the 3D pose of a person in real-time, which can be used in a variety of applications, such as virtual reality, human-computer interaction, and motion analysis.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in 3d-human-pose-estimation-1

Trend

Dataset

Best Model

Actions

Human3.6M

Human3.6M

3DPW

3DPW

MPI-INF-3DHP

MPI-INF-3DHP

Libraries

i

Use these libraries to find 3d-human-pose-estimation-1 models and implementations

open-mmlab/mmpose

10 papers 5,879

Datasets

Human3.6M

Waymo Open Dataset

Waymo Open Dataset

3DPW

AMASS

MPI-INF-3DHP

DensePose

Subtasks

Pose Prediction Monocular 3D Human Pose Estimation 3D Multi-Person Pose Estimation 3D human pose and shape estimation 3D human pose and shape estimation

Most implemented papers

Convolutional Pose Machines

Yaser Sheikh, S. Wei, V. Ramakrishna, T. Kanade•Fri Jan 29 2016

This work designs a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference in structured prediction tasks such as articulated pose estimation.

2870

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

HumanEva-I

HumanEva-I

H3WB

H3WB

Total Capture

Total Capture

AGORA

AGORA

EMDB

EMDB

Panoptic

Panoptic

Surreal

Surreal

3D Poses in the Wild Challenge

3D Poses in the Wild Challenge

AIST++

AIST++

RICH

RICH

SLOPER4D

SLOPER4D

UBody

UBody

SkiPose

SkiPose

Geometric Pose Affordance

Geometric Pose Affordance

DHP19

DHP19

SPEC-MTP

SPEC-MTP

CHALL H80K

CHALL H80K

Geometric Pose Affordance

Geometric Pose Affordance

JTA

JTA

HSPACE

HSPACE

3DOH50K

3DOH50K

Waymo Open Dataset

Waymo Open Dataset

ITOP front-view

ITOP front-view

3 papers 2,976

ailingzengzzz/Split-and-Recombine-N…

3 papers 40

3 papers 18

PaddlePaddle/PaddleDetection

2 papers 12,827

Daniil-Osokin/lightweight-human-pos…

2 papers 665

MandyMo/pytorch_HMR

2 papers 593

chingswy/HumanPoseMemo

2 papers 174

CAMMA-public/mvor

2 papers 54

DensePose

LSP

Panoptic

AGORA

TotalCapture

Weakly-supervised 3D Human Pose Estimation

Multi-Hypotheses 3D Human Pose Estimation

Egocentric Pose Estimation

3D Absolute Human Pose Estimation

Global 3D Human Pose Estimation

0

Deep High-Resolution Representation Learning for Human Pose Estimation

Ke Sun, Bin Xiao, Dong Liu, Jingdong Wang•Sun Feb 24 2019

This paper proposes a network that maintains high-resolution representations through the whole process of human pose estimation and empirically demonstrates the effectiveness of the network through the superior pose estimation results over two benchmark datasets: the COCO keypoint detection dataset and the MPII Human Pose dataset.

4672 0

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition

Yuanjun Xiong, Dahua Lin, Sijie Yan•Mon Jan 22 2018

A novel model of dynamic skeletons called Spatial-Temporal Graph Convolutional Networks (ST-GCN), which moves beyond the limitations of previous methods by automatically learning both the spatial and temporal patterns from data.

4837 0

DensePose: Dense Human Pose Estimation in the Wild

Iasonas Kokkinos, R. Güler, N. Neverova•Wed Jan 31 2018

This work establishes dense correspondences between an RGB image and a surface-based representation of the human body, a task referred to as dense human pose estimation, and improves accuracy through cascading, obtaining a system that delivers highly-accurate results at multiple frames per second on a single gpu.

1523 0

A Simple Yet Effective Baseline for 3d Human Pose Estimation

J. Little, Julieta Martinez, J. Romero, Rayat Hossain•Sun May 07 2017

The results indicate that a large portion of the error of modern deep 3d pose estimation systems stems from their visual analysis, and suggests directions to further advance the state of the art in 3d human pose estimation.

1446 0

3D Human Pose Estimation in Video With Temporal Convolutions and Semi-Supervised Training

Michael Auli, David Grangier, Dario Pavllo, Christoph Feichtenhofer•Tue Nov 27 2018

It is demonstrated that 3D poses in video can be effectively estimated with a fully convolutional model based on dilated temporal convolutions over 2D keypoints and back-projection, a simple and effective semi-supervised training method that leverages unlabeled video data is introduced.

1171 0

Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image

Chris Russell, L. Agapito, Denis Tomè•Sat Dec 31 2016

An integrated approach is taken that fuses probabilistic knowledge of 3D human pose with a multi-stage CNN architecture and uses the knowledge of plausible 3D landmark locations to refine the search for better 2D locations.

529 0

End-to-End Recovery of Human Shape and Pose

Angjoo Kanazawa, Jitendra Malik, Michael J. Black, D. Jacobs•Sun Dec 17 2017

This work introduces an adversary trained to tell whether human body shape and pose parameters are real or not using a large database of 3D human meshes, and produces a richer and more useful mesh representation that is parameterized by shape and 3D joint angles.

1967 0

BlazePose: On-device Real-time Body Pose tracking

Valentin Bazarevsky, Ivan Grishchenko, Karthik Raveendran, Tyler Lixuan Zhu, Fan Zhang, Matthias Grundmann•Tue Jun 16 2020

BlazePose is presented, a lightweight convolutional neural network architecture for human pose estimation that is tailored for real-time inference on mobile devices that uses both heatmaps and regression to keypoint coordinates.

735 0

MogaNet: Multi-order Gated Aggregation Network

Zicheng Liu, Jiangbin Zheng, Cheng Tan, Siyuan Li, Stan Z. Li, Zedong Wang, Haitao Lin, Di Wu, Zhiyuan Chen•Sun Nov 06 2022

MogaNet encapsulates conceptually simple yet effective convolutions and gated aggregation into a compact module, where discriminative features are efficiently gathered and contextualized adaptively and exhibits great scalability, impressive efficiency of parameters, and competitive performance compared to state-of-the-art ViTs and ConvNets on ImageNet and various downstream vision benchmarks.

130 0

Adding a benchmark result helps the community track progress.

3D Human Pose Estimation | State-of-the-Art