Literature DB >> 31388959

Fusing information from multiple 2D depth cameras for 3D human pose estimation in the operating room.

Lasse Hansen1, Marlin Siebert2, Jasper Diesel3, Mattias P Heinrich2.   

Abstract

PURPOSE: For many years, deep convolutional neural networks have achieved state-of-the-art results on a wide variety of computer vision tasks. 3D human pose estimation makes no exception and results on public benchmarks are impressive. However, specialized domains, such as operating rooms, pose additional challenges. Clinical settings include severe occlusions, clutter and difficult lighting conditions. Privacy concerns of patients and staff make it necessary to use unidentifiable data. In this work, we aim to bring robust human pose estimation to the clinical domain.
METHODS: We propose a 2D-3D information fusion framework that makes use of a network of multiple depth cameras and strong pose priors. In a first step, probabilities of 2D joints are predicted from single depth images. These information are fused in a shared voxel space yielding a rough estimate of the 3D pose. Final joint positions are obtained by regressing into the latent pose space of a pre-trained convolutional autoencoder.
RESULTS: We evaluate our approach against several baselines on the challenging MVOR dataset. Best results are obtained when fusing 2D information from multiple views and constraining the predictions with learned pose priors.
CONCLUSIONS: We present a robust 3D human pose estimation framework based on a multi-depth camera network in the operating room. Depth images as only input modalities make our approach especially interesting for clinical applications due to the given anonymity for patients and staff.

Entities:  

Keywords:  2D–3D information fusion; Convolutional autoencoder; Deep learning; Human pose estimation; Operating room

Mesh:

Year:  2019        PMID: 31388959     DOI: 10.1007/s11548-019-02044-7

Source DB:  PubMed          Journal:  Int J Comput Assist Radiol Surg        ISSN: 1861-6410            Impact factor:   2.924


  6 in total

1.  Statistical modeling and recognition of surgical workflow.

Authors:  Nicolas Padoy; Tobias Blum; Seyed-Ahmad Ahmadi; Hubertus Feussner; Marie-Odile Berger; Nassir Navab
Journal:  Med Image Anal       Date:  2010-12-08       Impact factor: 8.545

2.  Video recording of the operating room--is anonymity possible?

Authors:  Megan R Silas; Philippe Grassia; Alexander Langerman
Journal:  J Surg Res       Date:  2015-04-09       Impact factor: 2.192

3.  Temporal Trends and Characteristics of Reportable Health Data Breaches, 2010-2017.

Authors:  Thomas H McCoy; Roy H Perlis
Journal:  JAMA       Date:  2018-09-25       Impact factor: 56.272

4.  Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments.

Authors:  Catalin Ionescu; Dragos Papava; Vlad Olaru; Cristian Sminchisescu
Journal:  IEEE Trans Pattern Anal Mach Intell       Date:  2014-07       Impact factor: 6.226

5.  In-Bed Pose Estimation: Deep Learning With Shallow Dataset.

Authors:  Shuangjun Liu; Yu Yin; Sarah Ostadabbas
Journal:  IEEE J Transl Eng Health Med       Date:  2019-01-14       Impact factor: 3.316

6.  Patient-Specific Pose Estimation in Clinical Environments.

Authors:  Kenny Chen; Paolo Gabriel; Abdulwahab Alasfour; Chenghao Gong; Werner K Doyle; Orrin Devinsky; Daniel Friedman; Patricia Dugan; Lucia Melloni; Thomas Thesen; David Gonda; Shifteh Sattar; Sonya Wang; Vikash Gilja
Journal:  IEEE J Transl Eng Health Med       Date:  2018-10-10       Impact factor: 3.316

  6 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.