Graph-Based CNNs With Self-Supervised Module for 3D Hand Pose Estimation From Monocular RGB

Guo, Shaoxiang; Rigall, Eric; Qi, Lin; Dong, Xinghui; Li, Haiyan; Dong, Jiwen

doi:10.1109/tcsvt.2020.3004453

Cited by 32 publications

(17 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Features learned from these tasks can be transferred to image tasks with good performance. Guo et al designed a self-correction module to co-train networks in previous stages for hand pose estimation [43]. Xu et al proposed a set of pretext tasks specifically designed for sketches [44].…”

Section: A Intra-sample Learningmentioning

confidence: 99%

An Improved Inter-Intra Contrastive Learning Framework on Self-Supervised Video Representation

Wang

Yamasaki

2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

In this paper, we propose a self-supervised contrastive learning method to learn video feature representations. In traditional self-supervised contrastive learning methods, constraints from anchor, positive, and negative data pairs are used to train the model. In such a case, different samplings of the same video are treated as positives, and video clips from different videos are treated as negatives. Because the spatio-temporal information is important for video representation, we set the temporal constraints more strictly by introducing intra-negative samples. In addition to samples from different videos, negative samples are extended by breaking temporal relations in video clips from the same anchor video. With the proposed Inter-Intra Contrastive (IIC) framework, we can train spatio-temporal convolutional networks to learn feature representations from videos. Strong data augmentations, residual clips, as well as head projector are utilized to construct an improved version. Three kinds of intra-negative generation functions are proposed and extensive experiments using different network backbones are conducted on benchmark datasets. Without using pre-computed optical flow data, our improved version can outperform previous IIC by a large margin, such as 19.4% (from 36.8% to 56.2%) and 5.2% (from 15.5% to 20.7%) points improvements in top-1 accuracy on UCF101 and HMDB51 datasets for video retrieval, respectively. For video recognition, over 3% points improvements can also be obtained on these two benchmark datasets. Discussions and visualizations validate that our IICv2 can capture better temporal clues and indicate the potential mechanism.

show abstract

Section: A Intra-sample Learningmentioning

confidence: 99%

An Improved Inter-Intra Contrastive Learning Framework on Self-Supervised Video Representation

Wang

Yamasaki

2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…Compared to it, our method is more suitable for video tasks. Guo et al [32] use the 2D heat map as the intermediate supervised signal for 3D hand pose estimation. In comparison with this, our method fuses three self-supervised signals and makes avoiding collapse solution into consideration.…”

Section: A Learning From Video Contentmentioning

confidence: 99%

Self-Supervised Representation Learning for Videos by Segmenting via Sampling Rate Order Prediction

Huang

Wang

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Self-supervised representation learning for videos has been very attractive recently because these methods exploit the information inherently obtained from the video itself instead of annotated labels that is quite time-consuming. However, existing methods ignore the importance of global observation while performing spatio-temporal transformation perception, which highly limits the expression capabilities of the video representation. This paper proposes a novel pretext task that combines the temporal information perception of the video with the motion amplitude perception of moving objects to learn the spatiotemporal representation of the video. Specifically, given a video clip containing several video segments, each video segment is sampled by different sampling rates and the order of video segments is disrupted. Then, the network is used to regress the sampling rate of each video segment and classify the order of input video segments. In the pre-training stage, the network can learn rich spatio-temporal semantic information where content-related contrastive learning is introduced to make the learned video representation more discriminative. To alleviate the appearance dependency caused by contrastive learning, we design a novel and robust vector similarity measurement approach, which can take feature alignment into consideration. Moreover, a view synthesis framework is proposed to further improve the performance of contrastive learning by automatically generating reasonable transformed views. We conduct benchmark experiments with several 3D backbone networks on two datasets. The results show that our proposed method outperforms the existing state-of-theart methods across the three backbones on two downstream tasks of human action recognition and video retrieval.

show abstract

“…[25] exploits spatial and temporal relationships for 3D human and hand pose estimation tasks. [24] introduces a self-supervised module that uses 2D relationships and 3D geometric knowledge to reduce the gap between 2D and 3D spaces. [27] proposes a UNet-based GCNs to estimate the 3D hand pose and the 6D object pose.…”

Section: Related Workmentioning

confidence: 99%

“…Inspired by the natural graph representation of the hand, recent studies use graph convolution networks (GCN) [22] to model skeletal constraints between joints [23][24][25][26][27]. [25] exploits annotated video frames to enforce spatial and temporal relationships between the joints based on predefined semantic meanings.…”

Section: Introductionmentioning

confidence: 99%

A hybrid classification-regression approach for 3D hand pose estimation using graph convolutional networks

Kourbane¹,

Genç²

2021

Preprint

View full text Add to dashboard Cite

Hand pose estimation is a crucial part of a wide range of augmented reality and human-computer interaction applications. Predicting the 3D hand pose from a single RGB image is challenging due to occlusion and depth ambiguities. GCN-based (Graph Convolutional Networks) methods exploit the structural relationship similarity between graphs and hand joints to model kinematic dependencies between joints. These techniques use predefined or global learned joint relationships, which may fail to capture pose dependent constraints. To address this problem, we propose a two-stage GCN-based framework that learns per-pose relationship constraints. Specifically, the first phase quantizes the 2D/3D space to classify the joints into 2D/3D blocks based on their locality. This spatial dependency information guides this phase to estimate reliable 2D and 3D poses. The second stage further improves the 3D estimation through a GCN-based module that uses an adaptative nearest neighbor algorithm to determine joint relationships. Extensive experiments show that our multi-stage GCN approach yields an efficient model that produces accurate 2D/3D hand poses and outperforms the state-of-the-art on two public datasets.

show abstract

Graph-Based CNNs With Self-Supervised Module for 3D Hand Pose Estimation From Monocular RGB

Cited by 32 publications

References 32 publications

An Improved Inter-Intra Contrastive Learning Framework on Self-Supervised Video Representation

An Improved Inter-Intra Contrastive Learning Framework on Self-Supervised Video Representation

Self-Supervised Representation Learning for Videos by Segmenting via Sampling Rate Order Prediction

A hybrid classification-regression approach for 3D hand pose estimation using graph convolutional networks

Contact Info

Product

Resources

About