TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors
PLoS ONE · مج 19 (12) · e0310831-e0310831 · 2024
الملخّص
Accurate and flexible 3D pose estimation for virtual entities is a strenuous task in computer vision applications. Conventional methods struggle to capture realistic movements; thus, creative solutions that can handle the complexities of genuine avatar interactions in dynamic virtual environments are imperative. In order to tackle the problem of precise 3D pose estimation, this work introduces TRI-POSE-Net, a model intended for scenarios with limited supervision. The proposed technique, which is based on ResNet-50 and includes integrated Selective Kernel Network (SKNet) blocks, has proven to be efficient for feature extraction customised specifically to pose estimation scenarios. Furthermore, trifocal tensors and their trio-view geometry allow us to generate 3D ground truth poses from 2D poses, resulting in more refined triangulations. Through the proposed approach, the 3D poses can be estimated from a single 2D RGB image. Moreover, the proposed approach was evaluated on the HumanEva-I dataset yielding a Mean-Per-Joint-Position-Error (MPJPE) of 47.6 under self-supervision and an MPJPE of 29.9 under full supervision. In comparison with the other works, the proposed work has performed well in the self-supervision paradigm.
التفاصيل
- المجلة
- PLoS ONE
- الناشر
- Public Library of Science
- المجلد
- 19
- العدد
- 12
- الصفحات
- e0310831-e0310831
- الترقيم الدولي ISSN
- 1932-6203
- السنة
- 2024
- اللغة
- en
- الترخيص
- public-domain
- الوصول المفتوح
- gold
- الاستشهادات
- 3
- المصدر
- scopus
الكلمات المفتاحية
الموضوعات
Human Pose and Action Recognition (Computer Science) · Face recognition and analysis (Computer Science) · 3D Shape Modeling and Analysis (Engineering)
استشهد بهذا
Khan, N. A., Alarfaj, A. A., Alabdulqader, E. A., Zamzami, N., Umer, M., Innab, N., Kim, T. (2024). TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors. PLoS ONE, 19(12), e0310831-e0310831. https://doi.org/10.1371/journal.pone.0310831
Khan, N. A., Alarfaj, A. A., Alabdulqader, E. A., Zamzami, N., Umer, M., Innab, N., Kim, T. 2024, 'TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors', PLoS ONE, vol. 19(12), e0310831-e0310831. Available at: https://doi.org/10.1371/journal.pone.0310831
Khan NA, Alarfaj AA, Alabdulqader EA, Zamzami N, Umer M, Innab N, Kim T. TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors. PLoS ONE. 2024;19(12), e0310831-e0310831.
@article{khan20241040,
title = {TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors},
author = {Nabeel Ahmed Khan and Aisha Ahmed Alarfaj and Ebtisam Abdullah Alabdulqader and Nuha Zamzami and Muhammad Umer and Nisreen Innab and Tai-hoon Kim},
year = {2024},
journal = {PLoS ONE},
volume = {19},
number = {12},
pages = {e0310831-e0310831},
publisher = {Public Library of Science},
doi = {10.1371/journal.pone.0310831},
url = {https://doi.org/10.1371/journal.pone.0310831}
}TY - JOUR AU - Nabeel Ahmed Khan AU - Aisha Ahmed Alarfaj AU - Ebtisam Abdullah Alabdulqader AU - Nuha Zamzami AU - Muhammad Umer AU - Nisreen Innab AU - Tai-hoon Kim TI - TRI-POSE-Net: Adaptive 3D human pose estimation through selective kernel networks and self-supervision with trifocal tensors PY - 2024 JO - PLoS ONE VL - 19 IS - 12 SP - e0310831-e0310831 DO - 10.1371/journal.pone.0310831 UR - https://doi.org/10.1371/journal.pone.0310831 ER -
