3D hand reconstruction from a single image based on biomechanical constraints

Li, Guiqing; Wu, Zihui; Liu, Yuxin; Zhang, Huiqian; Nie, Yongwei; Mao, Aihua

doi:10.1007/s00371-021-02250-y

Cited by 6 publications

(2 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some works such as Cai et al ( 2019 ) used refinement models to adjust the poses with limits and rules. However, most of these works (Ryf and Weymann, 1995 ; Cobos et al, 2008 ; Chen Chen et al, 2013 ; Melax et al, 2013 ; Sridhar et al, 2013 ; Xu and Cheng, 2013 ; Tompson et al, 2014 ; Aristidou, 2018 ; Li et al, 2021 ) apply the rules and bounds after estimating the pose of the hand using post-processing methods such as inverse kinematics and bound penalization. Recent works used biomechanical constraints for hand pose estimation using 2D images in the neural network's cost function to penalize the joints.…”

Section: Related Workmentioning

confidence: 99%

“…Hence there are poses where the joint angles exceed the anatomical bounds. Li et al ( 2021 ) used a model-based iterative approach by first applying the PoseNet (Choi et al, 2020 ) and then computing the motion parameters. The drawback of this approach is that it depends on the PoseNet for recovering the primary joint positions and fails to operate if PoseNet fails to predict the pose.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Single Shot Corrective CNN for Anatomically Correct 3D Hand Pose Estimation

Isaac¹,

Manivannan²,

Ravindran³

2022

Front. Artif. Intell.

View full text Add to dashboard Cite

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.

show abstract