Spatial Transformation Based 2D Image Segmentation for Single Human Pose Estimation with Improved Efficiency
Main Article Content
Abstract
Human pose evaluation on multi-person is challenging than human pose evaluation on single-person . Even when the human detectors have given their best results, small errors are available in localization part and recognition. These errors in detection lead to failures in human pose estimation. So, we propose an architecture for human pose evaluation although when the errors are present in the human detection. The model consists of two main steps: first step is to detect the human and next step is to evaluate the human pose for the detected human. The components used primarily for human pose estimation is the STN(Spatial Transformer Network), SPPE(Single Person Pose Estimator) and Pose-NMS(Non-Maximum Suppression). This method is tested on the MPII(multi-person) dataset.
Downloads
Metrics
Article Details
Licensing
TURCOMAT publishes articles under the Creative Commons Attribution 4.0 International License (CC BY 4.0). This licensing allows for any use of the work, provided the original author(s) and source are credited, thereby facilitating the free exchange and use of research for the advancement of knowledge.
Detailed Licensing Terms
Attribution (BY): Users must give appropriate credit, provide a link to the license, and indicate if changes were made. Users may do so in any reasonable manner, but not in any way that suggests the licensor endorses them or their use.
No Additional Restrictions: Users may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.