Transformer for skeleton-based action recognition: A review of recent advances

W **n, R Liu, Y Liu, Y Chen, W Yu, Q Miao - Neurocomputing, 2023‏ - Elsevier
Skeleton-based action recognition has rapidly become one of the most popular and
essential research topics in computer vision. The task is to analyze the characteristics of …

Deep contrastive representation learning with self-distillation

Z **ao, H **ng, B Zhao, R Qu, S Luo… - … on Emerging Topics …, 2023‏ - ieeexplore.ieee.org
Recently, contrastive learning (CL) is a promising way of learning discriminative
representations from time series data. In the representation hierarchy, semantic information …

On the use of deep learning for video classification

A Ur Rehman, SB Belhaouari, MA Kabir, A Khan - Applied Sciences, 2023‏ - mdpi.com
The video classification task has gained significant success in the recent years. Specifically,
the topic has gained more attention after the emergence of deep learning models as a …

Dynamic dense graph convolutional network for skeleton-based human motion prediction

X Wang, W Zhang, C Wang, Y Gao… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Graph Convolutional Networks (GCN) which typically follows a neural message passing
framework to model dependencies among skeletal joints has achieved high success in …

Self-supervised graph-level representation learning with adversarial contrastive learning

X Luo, W Ju, Y Gu, Z Mao, L Liu, Y Yuan… - ACM Transactions on …, 2023‏ - dl.acm.org
The recently developed unsupervised graph representation learning approaches apply
contrastive learning into graph-structured data and achieve promising performance …

Neighbor-guided consistent and contrastive learning for semi-supervised action recognition

J Wu, W Sun, T Gan, N Ding, F Jiang… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Semi-supervised learning has been well established in the area of image classification but
remains to be explored in video-based action recognition. FixMatch is a state-of-the-art semi …

Self-supervised 3D action representation learning with skeleton cloud colorization

S Yang, J Liu, S Lu, EM Hwa, Y Hu… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
3D Skeleton-based human action recognition has attracted increasing attention in recent
years. Most of the existing work focuses on supervised learning which requires a large …

Msvit: training multiscale vision transformers for image retrieval

X Li, J Yu, S Jiang, H Lu, Z Li - IEEE Transactions on Multimedia, 2023‏ - ieeexplore.ieee.org
The recently developed vision transformer (ViT) has achieved promising results on image
retrieval compared to convolutional neural networks. However, most of these vision …

Individual and structural graph information bottlenecks for out-of-distribution generalization

L Yang, J Zheng, H Wang, Z Liu… - … on Knowledge and …, 2023‏ - ieeexplore.ieee.org
Out-of-distribution (OOD) graph generalization are critical for many real-world applications.
Existing methods neglect to discard spurious or noisy features of inputs, which are irrelevant …

Sptr: Structure-preserving transformer for unsupervised indoor depth completion

L Zhao, W Zheng, Y Duan, J Zhou… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Recovering a dense depth map from a pair of indoor RGB and sparse depth images in an
unsupervised manner is paramount in applications such as autonomous driving and 3D …