Summary of Learning Mutual Excitation For Hand-to-hand and Human-to-human Interaction Recognition, by Mengyuan Liu et al.
Learning Mutual Excitation for Hand-to-Hand and Human-to-Human Interaction Recognition
by Mengyuan Liu, Chen Chen, Songtao Wu, Fanyang Meng, Hong Liu
First submitted to arxiv on: 4 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 proposed mutual excitation graph convolutional network (me-GCN) tackles interactive action recognition by stacking mutual excitation graph convolution (me-GC) layers. It extracts adjacency matrices from individual entities, models their mutual constraints, and merges deep features from pairwise entities. Compared to traditional graph convolution, me-GCN learns mutual information in each layer and stage of operations. This approach outperforms state-of-the-art GCN-based and Transformer-based methods on hand-to-hand interaction (Assembely101) and human-to-human interaction datasets (NTU60-Interaction and NTU120-Interaction). |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers developed a new way to recognize actions that involve multiple people or objects interacting with each other. They created a special kind of computer network called the mutual excitation graph convolutional network, which is better at recognizing these types of interactions than previous methods. This new approach was tested on several large datasets and performed better than others in identifying various human-to-human and hand-to-hand interactions. |
Keywords
* Artificial intelligence * Convolutional network * Gcn * Transformer
