Abstract
Hand pose estimation in 3D from depth images is a highly complex task. Current state-of-the-art 3D hand pose estimators focus only on the accuracy of the model as measured by how closely it matches the ground truth hand pose but overlook the resulting hand pose's anatomical correctness. In this paper, we present the Single Shot Corrective CNN (SSC-CNN) to tackle the problem of enforcing anatomical correctness at the architecture level. In contrast to previous works which use post-facto pose filters, SSC-CNN predicts the hand pose that conforms to the human hand's biomechanical bounds and rules in a single forward pass. The model was trained and tested on the HANDS2017 and MSRA datasets. Experiments show that our proposed model shows comparable accuracy to the state-of-the-art models as measured by the ground truth pose. However, the previous methods have high anatomical errors, whereas our model is free from such errors. Experiments show that our proposed model shows zero anatomical errors along with comparable accuracy to the state-of-the-art models as measured by the ground truth pose. The previous methods have high anatomical errors, whereas our model is free from such errors. Surprisingly even the ground truth provided in the existing datasets suffers from anatomical errors, and therefore Anatomical Error Free (AEF) versions of the datasets, namely AEF-HANDS2017 and AEF-MSRA, were created.
Highlights
Hand pose estimation in 3D is the task of predicting the pose of the hand in 3D space provided the depth image of the hand
Our paper provides the following contributions: (1) A novel approach by incorporating biomechanical filter functions in the model architecture, (2) a hand pose estimator that guarantees zero anatomical error while maintaining low deviation from the ground truth pose, (3) we show that these anatomical rules and bounds were not maintained when creating the HANDS2017 and the MSRA hand datasets, and (4) an Anatomical Error Free (AEF) version of the datasets called AEF-HANDS2017 and AEF-MSRA was created
The wrist joint is the root of the hand and is simplified to 6 degrees of freedom (DoF) as it is the result of the chain of movements from the shoulder to the arm
Summary
Hand pose estimation in 3D is the task of predicting the pose of the hand in 3D space provided the depth (or 2D) image of the hand. The main metric used for comparison in this paper is the anatomical error of the hand pose, which is computed by using the joint angles measured for each joint in the hand pose after prediction Our paper provides the following contributions: (1) A novel approach by incorporating biomechanical filter functions in the model architecture, (2) a hand pose estimator that guarantees zero anatomical error while maintaining low deviation from the ground truth pose, (3) we show that these anatomical rules and bounds were not maintained when creating the HANDS2017 and the MSRA hand datasets, and (4) an Anatomical Error Free (AEF) version of the datasets called AEF-HANDS2017 and AEF-MSRA was created.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.