computer-vision-7

Hand Pose Estimation

3260 papers • 126 benchmarks • 313 datasets

Hand pose estimation is the task of finding the joints of the hand from an image or set of video frames. ( Image credit: Pose-REN )

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in hand-pose-estimation-14

Trend

Dataset

Best Model

Actions

NYU Hands

ICVL Hands

MSRA Hands

Libraries

i

Use these libraries to find hand-pose-estimation-14 models and implementations

open-mmlab/mmpose

3 papers 6,026

Datasets

RHD

Subtasks

3D Hand Pose Estimation

Most implemented papers

Learning to Estimate 3D Hand Pose from Single RGB Images

T. Brox, Christiane Zimmermann•Tue May 02 2017

A deep network is proposed that learns a network-implicit 3D articulation prior that yields good estimates of the 3D pose from regular RGB images, and a large scale 3D hand pose dataset based on synthetic hand models is introduced.

774

Content

HANDS 2017

HANDS 2019

COCO-WholeBody

K2HPD

Custom FINNgers

3DPW

ICVL

patiencefromzhou/simplehand

2 papers 77

rongakowang/densemutualattention

2 papers 39

NYU Hand

First-Person Hand Action Benchmark

ARCTIC

MSRA Hand

0

Paper Graph

Learning from Simulated and Unsupervised Images through Adversarial Training

Tomas Pfister, A. Shrivastava, Oncel Tuzel, J. Susskind, Wenda Wang, Russ Webb•Wed Dec 21 2016

This work develops a method for S+U learning that uses an adversarial network similar to Generative Adversarial Networks (GANs), but with synthetic images as inputs instead of random vectors, and makes several key modifications to the standard GAN algorithm to preserve annotations, avoid artifacts, and stabilize training.

1869 0

Paper Graph

V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map

Kyoung Mu Lee, Gyeongsik Moon, Ju Yong Chang•Sun Nov 19 2017

This model is designed as a 3D CNN that provides accurate estimates while running in real-time and outperforms previous methods in almost all publicly available 3D hand and human pose estimation datasets and placed first in the HANDS 2017 frame-based3D hand pose estimation challenge.

451 0

Paper Graph

HOnnotate: A Method for 3D Annotation of Hand and Object Poses

Markus Oberweger, V. Lepetit, Shreyas Hampali, Mahdi Rad•Mon Jul 01 2019

We propose a method for annotating images of a hand manipulating an object with the 3D poses of both the hand and the object, together with a dataset created using this method. Our motivation is the current lack of annotated real images for this problem, as estimating the 3D poses is challenging, mostly because of the mutual occlusions between the hand and the object. To tackle this challenge, we capture sequences with one or several RGB-D cameras and jointly optimize the 3D hand and object poses over all the frames simultaneously. This method allows us to automatically annotate each frame with accurate estimates of the poses, despite large mutual occlusions. With this method, we created HO-3D, the first markerless dataset of color images with 3D annotations for both the hand and object. This dataset is currently made of 77,558 frames, 68 sequences, 10 persons, and 10 objects. Using our dataset, we develop a single RGB image-based method to predict the hand pose when interacting with objects under severe occlusions and show it generalizes to objects not seen in the dataset.

451 0

Paper Graph

Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wild

M. Bronstein, S. Zafeiriou, Iasonas Kokkinos, R. Güler, Dominik Kulon•Fri Apr 03 2020

This work introduces a simple and effective network architecture for monocular 3D hand pose estimation consisting of an image encoder followed by a mesh convolutional decoder that is trained through a direct3D hand mesh reconstruction loss.

231 0

Paper Graph

DeepPrior++: Improving Fast and Accurate 3D Hand Pose Estimation

Markus Oberweger, V. Lepetit•Sun Aug 27 2017

With simple improvements: adding ResNet layers, data augmentation, and better initial hand localization, DeepPrior achieves better or similar performance than more sophisticated recent methods on the three main benchmarks (NYU, ICVL, MSRA) while keeping the simplicity of the original method.

251 0

Paper Graph

Learning Pose Specific Representations by Predicting Different Views

H. Bischof, Georg Poier, David Schinagl•Mon Apr 09 2018

A method to learn representations, which are very specific for articulated poses, without the need for labeled training data, that consistently surpasses the performance of its fully supervised counterpart, while reducing the amount of needed labeled samples by at least one order of magnitude.

26 0

Paper Graph

End-to-End Hand Mesh Recovery From a Monocular RGB Image

Xiong Zhang, Wen Zheng, Qiang Li, Wenbo Zhang•Sun Feb 24 2019

Qualitative experiments show that the HAMR framework is capable of recovering appealing 3D hand mesh even in the presence of severe occlusions, and outperforms the state-of-the-art methods for both 2D and3D hand pose estimation from a monocular RGB image on several benchmark datasets.

237 0

Paper Graph

3D Hand Shape and Pose Estimation From a Single RGB Image

Jianfei Cai, Junsong Yuan, Yuncheng Li, Liuhao Ge, Zhou Ren, Zehao Xue, Yingying Wang•Sat Mar 02 2019

This work proposes a Graph Convolutional Neural Network (Graph CNN) based method to reconstruct a full 3D mesh of hand surface that contains richer information of both 3D hand shape and pose and proposes a weakly-supervised approach by leveraging the depth map as a weak supervision in training.

487 0

Paper Graph

Adding a benchmark result helps the community track progress.