Fast and Accurate 3D Hand Pose Estimation via Recurrent Neural Network for Capturing Hand Articulations

Yoo, Cheol Hwan; Ji, Seo-Won; Shin, Yong-Goo; Kim, Seung-Wook; Ko, Sung Jea

doi:10.1109/access.2020.3001637

Cited by 14 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of the proposed method was compared with previous state-of-the-art methods using a 2D or 3D depth image input. The 3D input models are a 3DCNN [16], SHPR-Net [29], HandPointNet [30], Point-to-Point [31], and V2V-PoseNet [14], and the 2D input models are a DeepModel [19], Deep-Prior [12], Deep-Prior++ [13], Feedback [32], REN-4x6x6 [25], REN-9x6x6 [33], Pose-REN [17], CrossInfoNet [28], HCRNN [34], and A2J [35].…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 99%

3D Hand Pose Estimation via Graph-Based Reasoning

Song

Kang

2021

IEEE Access

View full text Add to dashboard Cite

Hand pose estimation from a single depth image has recently received significant attention owing to its importance in many applications requiring human-computer interaction. The rapid progress of convolutional neural networks (CNNs) and technological advances in low-cost depth cameras have greatly improved the performance of the hand pose estimation method. Nevertheless, regressing joint coordinates is still a challenging task due to joint flexibility and self-occlusion. Previous hand pose estimation methods have limitations in relying on a deep and complex network structure without fully utilizing hand joint connections. A hand is an articulated object and consists of six parts that represent the palm and five fingers. The kinematic constraints can be obtained by modeling the dependency between adjacent joints. This paper proposes a novel CNN-based approach incorporating hand joint connections to features through both a global relation inference for the entire hand and local relation inference for each finger. Modeling the relations between the hand joints can alleviate critical problems for occlusion and self-similarity. We also present a hierarchical structure with six branches that independently estimate the position of the palm and five fingers by adding hand connections of each joint using graph reasoning based on graph convolutional networks. Experimental results on public hand pose datasets show that the proposed method outperforms previous state-of-the-art methods. Specifically, our method achieves the best accuracy compared to state-of-the-art methods on public datasets. In addition, the proposed method can be utilized for real-time applications with an execution speed of 103 fps in a single GPU environment.INDEX TERMS 3D hand pose estimation, depth image, graph convolutional network.

show abstract

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 99%

3D Hand Pose Estimation via Graph-Based Reasoning

Song

Kang

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Using these synthetic datasets, we can also compare the spectrum of poses covered by the currently available datasets and hence cover a broader spectrum of poses for training. Analyzing the history of the hands' motion using methods such as recurrent neural networks (Yoo et al, 2020 ) instead of processing only one instance of the hand can avoid erratic motions during self-occlusions and will be investigated in another study for adding the feature to the SSC-CNN. The history can include the velocity and acceleration of the joint motions, which also have biomechanical bounds and further enhance the pose realism during hand motion tracking.…”

Section: Conclusion Limitations and Future Workmentioning

confidence: 99%

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

“…More recent approaches to hand pose estimation from a single depth map decomposes the task into several sub-tasks (for palm and fingers) [49], [50]. In [49], the authors adopt two-branch cross-connection structure to share the beneficial complementary information between the sub-tasks.…”

Section: Related Workmentioning

confidence: 99%