Summary of Integrating Features For Recognizing Human Activities Through Optimized Parameters in Graph Convolutional Networks and Transformer Architectures, by Mohammad Belal (1) et al.
Integrating Features for Recognizing Human Activities through Optimized Parameters in Graph Convolutional Networks and Transformer Architectures
by Mohammad Belal, Taimur Hassan, Abdelfatah Hassan, Nael Alsheikh, Noureldin Elhendawi, Irfan Hussain
First submitted to arxiv on: 29 Aug 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Robotics (cs.RO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a study on human activity recognition using deep learning techniques, focusing on the influence of feature fusion on accuracy. The authors employ four publicly available datasets (HuGaDB, PKU-MMD, LARa, and TUG) to evaluate the performance of two deep learning models: Transformer and Parameter-Optimized Graph Convolutional Network (PO-GCN). The results show that PO-GCN outperforms standard models in activity recognition, with improvements in accuracy and F1-score on three out of four datasets. This study demonstrates the potential of feature fusion techniques in enhancing the performance of deep learning models for human activity recognition. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about using computers to recognize different human activities, like walking or sitting. It tries to improve how well these computers do this by combining information from two different types of computer models: Transformer and PO-GCN. The researchers use four different sets of data (HuGaDB, PKU-MMD, LARa, and TUG) to test their idea. They find that using both models together can make the computers better at recognizing activities. |
Keywords
» Artificial intelligence » Activity recognition » Convolutional network » Deep learning » F1 score » Gcn » Transformer
