Published on Tue Oct 13 2020

A review of 3D human pose estimation algorithms for markerless motion capture

Yann Desmarais, Denis Mottet, Pierre Slangen, Philippe Montesinos

Human pose estimation (HPE) in 3D is an active research field. In the last five years markerless motion captures techniques have seen their average error decrease from more than 10cm to less than 2cm.

0
0
0
Abstract

Human pose estimation (HPE) in 3D is an active research field that have many applications in entertainment, health and sport science, robotics. In the last five years markerless motion captures techniques have seen their average error decrease from more than 10cm to less than 2cm today. This evolution is mainly driven by the improvements in 2D pose estimation task that benefited from the use of convolutional networks. However with the multiplication of different approaches it can be difficult to identify what is more adapted to the specifics of any applications. We suggest to classify existing methods with a taxonomy based on the performance criteria of accuracy, speed and robustness. We review more than twenty methods from the last three years. Additionally we analyze the metrics, benchmarks and structure of the different pose estimation systems and propose several direction for future research. We hope to offer a good introduction to 3D markerless pose estimation as well as discussing the leading contemporary algorithms.

Thu Dec 24 2020
Computer Vision
Deep Learning-Based Human Pose Estimation: A Survey
0
0
0
Thu Aug 20 2020
Computer Vision
PhysCap: Physically Plausible Monocular 3D Motion Capture in Real Time
PhysCap is the first algorithm for physically plausible, real-time and marker-less human 3D motion capture with a single colour camera at 25 fps. Our algorithm first captures 3Dhuman poses purely kinematically. Our method captures physically plausible andtemporally stable global 3D human motion. from video in real time and in general scenes.
0
0
0
Mon May 03 2021
Computer Vision
Neural Monocular 3D Human Motion Capture with Physical Awareness
Physionical approach is aware of physical and environmental constraints. It combines a proportional-derivative controller with gains predicted by a neural network. The inputs to our system are 2D joint keypoints, which are canonicalised.
0
0
0
Sat Aug 04 2018
Computer Vision
Rethinking Pose in 3D: Multi-stage Refinement and Recovery for Markerless Motion Capture
CNN-based approach for multi-camera markerless motion capture of the human body. Unlike existing methods that first perform pose estimation on individual cameras and generate 3D models as post-processing, our approach makes use of 3D reasoning throughout a multi-stage approach.
0
0
0
Wed Dec 05 2018
Computer Vision
Capture Dense: Markerless Motion Capture Meets Dense Pose Estimation
We present a method to combine markerless motion capture and dense pose feature estimation into a single framework. We demonstrate that dense pose information can help for multiview/single-view motion capture. We improve the performance of available dense pose detector data by using multiv view motion capture data.
0
0
0
Tue Jun 02 2020
Computer Vision
Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
Vision-based monocular human pose estimation is one of the most fundamental and challenging problems in computer vision. The recent developments of deep learning techniques have brought significant progress and remarkable breakthroughs. This survey extensively reviews the recent deep learning-based 2D and 3D human poses estimation methods.
0
0
0