Constraint-Based Optimized Human Skeleton Extraction from Single-Depth Camera.

Sensors (Basel)

Institute of Computer Science II, University of Bonn, Endenicher Allee 19a, 53115 Bonn, Germany.

Published: June 2019

AI Article Synopsis

  • The paper addresses the challenges of accurately extracting human skeletons from single-depth cameras due to issues like occlusions, appearance variations, and sensor noise.
  • The authors propose a new method that incorporates human skeleton length conservation and symmetry constraints along with temporal data to improve the consistency and accuracy of human pose estimates from initial joint positions provided by existing SDKs.
  • Their approach is device-independent and can refine skeleton extraction in various SDKs, effectively detecting errors and predicting new joint locations, with experimental results showing its effectiveness and robustness.

Article Abstract

As a cutting-edge research topic in computer vision and graphics for decades, human skeleton extraction from single-depth camera remains challenging due to possibly occurring occlusions of different body parts, huge appearance variations, and sensor noise. In this paper, we propose to incorporate human skeleton length conservation and symmetry priors as well as temporal constraints to enhance the consistency and continuity for the estimated skeleton of a moving human body. Given an initial estimation of the skeleton joint positions provided per frame by the Kinect SDK or Nuitrack SDK, which do not follow the aforementioned priors and can prone to errors, our framework improves the accuracy of these pose estimates based on the length and symmetry constraints. In addition, our method is device-independent and can be integrated into skeleton extraction SDKs for refinement, allowing the detection of outliers within the initial joint location estimates and predicting new joint location estimates following the temporal observations. The experimental results demonstrate the effectiveness and robustness of our approach in several cases.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6603546PMC
http://dx.doi.org/10.3390/s19112604DOI Listing

Publication Analysis

Top Keywords

human skeleton
12
skeleton extraction
12
extraction single-depth
8
single-depth camera
8
joint location
8
location estimates
8
skeleton
6
constraint-based optimized
4
human
4
optimized human
4

Similar Publications

Large-scale genetic mapping for human brain asymmetry.

Handb Clin Neurol

March 2025

Language & Genetics Department, Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands; Department of Cognitive Neuroscience & Donders Community for Medical Neuroscience, Radboud University Medical Center, Nijmegen, The Netherlands. Electronic address:

Left-right asymmetry is an important aspect of human brain organization for functions including language and hand motor control, which can be altered in some psychiatric traits. The last 5 years have seen rapid advances in the identification of specific genes linked to variation in asymmetry of the human brain and/or handedness. These advances have been driven by a new generation of large-scale genome-wide association studies, carried out in samples ranging from roughly 16,000 to over 1.

View Article and Find Full Text PDF

ZNF667 alleviates the inflammatory damage in intervertebral disc degeneration via inhibiting NF-κB signaling pathway.

Zhong Nan Da Xue Xue Bao Yi Xue Ban

October 2024

Department of Spine Surgery, Second Xiangya Hospital, Central South University, Changsha 410011, China.

Objectives: With the aging population, the incidence of intervertebral disc degeneration (IDD) is increasing every year. The pathogenesis of IDD is complex, and there are currently no effective treatment options. This study aims to investigate the specific function and underlying mechanism of zinc finger protein 667 (ZNF667) in the inflammatory damage of nucleus pulposus cells in IDD.

View Article and Find Full Text PDF

Structurally diverse sesquiterpenes with anti-leukemia activity from the rare medicinal plant Liriodendron chinense.

Bioorg Chem

March 2025

Department of Natural Medicine, School of Pharmacy, Fudan University, Shanghai 201203, PR China. Electronic address:

A further phytochemical investigation on the branches and leaves of Liriodendron chinense, a rare medicinal and ornamental plant endemic to China, yielded thirty-four sesquiterpenes with diverse skeletons. The isolated compounds comprise 12 new naturally occurring sesquiterpenoids (1-12) and 22 known analogues (13-34). The new structures were elucidated by spectroscopic data analyses, and their absolute configurations (except for 10) were determined by electronic circular dichroism spectra, quantum-chemical calculations, and X-ray crystallography.

View Article and Find Full Text PDF

ObjectiveThis study aimed to assess the practicality and optimal approach for inserting an anterior occipital condyle screw, as well as to measure the screw placement characteristics.MethodsA total of 80 normal head and cervical spine computed tomography scans (40 males/40 females) were used to construct three-dimensional models. The average age of the participants was 45.

View Article and Find Full Text PDF

Pituitary neuroendocrine tumors (PitNETS) are common intracranial tumors, but extrasellar or ectopic PitNETS are very rare and supposed to originate from some pituitary remnants. They are mostly found in sphenoidal sinus. But particularly, ectopic clival PitNETS are highly aggressive and can cause bone invasion and can be misdiagnosed as other lesions of the skull base such as chordomas.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!