Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera
Recovering 3D structures from the monocular image sequence is an inherently ambiguous problem that has attracted considerable attention from several research communities. To resolve the ambiguities, a variety of additional priors, such as low-rank shape basis, have been proposed. In this paper, we m...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/17/9/2019 |
| _version_ | 1828152038816481280 |
|---|---|
| author | Xuan Wang Fei Wang Yanan Chen |
| author_facet | Xuan Wang Fei Wang Yanan Chen |
| author_sort | Xuan Wang |
| collection | DOAJ |
| description | Recovering 3D structures from the monocular image sequence is an inherently ambiguous problem that has attracted considerable attention from several research communities. To resolve the ambiguities, a variety of additional priors, such as low-rank shape basis, have been proposed. In this paper, we make two contributions. First, we introduce an assumption that 3D structures lie on the union of nonlinear subspaces. Based on this assumption, we propose a Non-Rigid Structure from Motion (NRSfM) method with kernelized low-rank representation. To be specific, we utilize the soft-inextensibility constraint to accurately recover 3D human motions. Second, we extend this NRSfM method to the marker-less 3D human pose estimation problem by combining with Convolutional Neural Network (CNN) based 2D human joint detectors. To evaluate the performance of our methods, we apply our marker-based method on several sequences from Utrecht Multi-Person Motion (UMPM) benchmark and CMU MoCap datasets, and then apply the marker-less method on the Human3.6M datasets. The experiments demonstrate that the kernelized low-rank representation is more suitable for modeling the complex deformation and the method consequently yields more accurate reconstructions. Benefiting from the CNN-based detector, the marker-less approach can be applied to more real-life applications. |
| first_indexed | 2024-04-11T22:06:03Z |
| format | Article |
| id | doaj.art-6837fa6fc15f4e84b50b551100a84997 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T22:06:03Z |
| publishDate | 2017-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6837fa6fc15f4e84b50b551100a849972022-12-22T04:00:43ZengMDPI AGSensors1424-82202017-09-01179201910.3390/s17092019s17092019Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB CameraXuan Wang0Fei Wang1Yanan Chen2The Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, No.28 Xianning West Road, Xi’an 710048, ChinaThe Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, No.28 Xianning West Road, Xi’an 710048, ChinaThe Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, No.28 Xianning West Road, Xi’an 710048, ChinaRecovering 3D structures from the monocular image sequence is an inherently ambiguous problem that has attracted considerable attention from several research communities. To resolve the ambiguities, a variety of additional priors, such as low-rank shape basis, have been proposed. In this paper, we make two contributions. First, we introduce an assumption that 3D structures lie on the union of nonlinear subspaces. Based on this assumption, we propose a Non-Rigid Structure from Motion (NRSfM) method with kernelized low-rank representation. To be specific, we utilize the soft-inextensibility constraint to accurately recover 3D human motions. Second, we extend this NRSfM method to the marker-less 3D human pose estimation problem by combining with Convolutional Neural Network (CNN) based 2D human joint detectors. To evaluate the performance of our methods, we apply our marker-based method on several sequences from Utrecht Multi-Person Motion (UMPM) benchmark and CMU MoCap datasets, and then apply the marker-less method on the Human3.6M datasets. The experiments demonstrate that the kernelized low-rank representation is more suitable for modeling the complex deformation and the method consequently yields more accurate reconstructions. Benefiting from the CNN-based detector, the marker-less approach can be applied to more real-life applications.https://www.mdpi.com/1424-8220/17/9/20193D human pose estimationmonocular reconstructionnon-rigid structure from motionkernel low-rank representation |
| spellingShingle | Xuan Wang Fei Wang Yanan Chen Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera Sensors 3D human pose estimation monocular reconstruction non-rigid structure from motion kernel low-rank representation |
| title | Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera |
| title_full | Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera |
| title_fullStr | Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera |
| title_full_unstemmed | Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera |
| title_short | Capturing Complex 3D Human Motions with Kernelized Low-Rank Representation from Monocular RGB Camera |
| title_sort | capturing complex 3d human motions with kernelized low rank representation from monocular rgb camera |
| topic | 3D human pose estimation monocular reconstruction non-rigid structure from motion kernel low-rank representation |
| url | https://www.mdpi.com/1424-8220/17/9/2019 |
| work_keys_str_mv | AT xuanwang capturingcomplex3dhumanmotionswithkernelizedlowrankrepresentationfrommonocularrgbcamera AT feiwang capturingcomplex3dhumanmotionswithkernelizedlowrankrepresentationfrommonocularrgbcamera AT yananchen capturingcomplex3dhumanmotionswithkernelizedlowrankrepresentationfrommonocularrgbcamera |