WYWei Yang

Wei Yang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Wei Yang has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Recognizing A Complex Human Behaviour via A Shallow Neural Network with Zero Video Training Sample

    by Qianglai Xie, Wei Lu, Wei Yang, Keyun Xiong, Lei Zhang, Leiyue Yao

    In contrast to human action recognition (HAR), understanding complex human behaviour (CHB), consisting of multiple basic actions, poses a significant challenge for researchers due to its extended duration, numerous types, and substantial data-labeling expenses. In this paper, a new approach to recognize CHB from a semantic point of view is proposed, which can be roughly summarized as judging by action quantization and action combination similarity. To fully evaluate the effectiveness of our method, the self-collected dataset – HanYue Action3D is extended to become the first public skeleton-based dataset with complex behavior samples and temporal calibration. Experimental results have demonstrated the feasibility and universal superiority of our method. Moreover, our method’s zero-shot learning capability bridges the divide between laboratory settings and real-world applications.


  • Open Access

    A Data Augmentation Method for Skeleton-Based Action Recognition with Relative Features

    by Junjie Chen, Wei Yang, Chenqi Liu, Leiyue Yao

    In recent years, skeleton-based human action recognition (HAR) approaches using convolutional neural network (CNN) models have made tremendous progress in computer vision applications. However, using relative features to depict human actions, in addition to preventing overfitting when the CNN model is trained on a few samples, is still a challenge. In this paper, a new motion image is introduced to transform spatial-temporal motion information into image-based representations. For each skeleton sequence, three relative features are extracted to describe human actions. The three relative features are consisted of relative coordinates, immediate displacement, and immediate motion orientation. In particular, the relative coordinates introduced in our paper not only depict the spatial relations of human skeleton joints but also provide long-term temporal information. To address the problem of small sample sizes, a data augmentation strategy consisting of three simple but effective data augmentation methods is proposed to expand the training samples. Because the generated color images are small in size, a shallow CNN model is suitable to extract the deep features of the generated motion images. Two small-scale but challenging skeleton datasets were used to evaluate the method, scoring 96.59% and 97.48% on the Florence 3D Actions dataset and UTkinect-Action 3D dataset, respectively. The results show that the proposed method achieved a competitive performance compared with the state-of-the-art methods. Furthermore, the augmentation strategy proposed in this paper effectively solves the overfitting problem and can be widely adopted in skeleton-based action recognition.