Summary of Local and Global Feature Attention Fusion Network For Face Recognition, by Wang Yu et al.
Local and Global Feature Attention Fusion Network for Face Recognition
by Wang Yu, Wei Wei
First submitted to arxiv on: 25 Nov 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 Local and Global Feature Attention Fusion (LGAF) network tackles the challenge of recognizing low-quality face images by adaptively allocating attention between local and global features based on feature quality. The LGAF network combines the strengths of local region similarity and global features to obtain more discriminative and high-quality face features. Additionally, the Multi-Head Multi-Scale Local Feature Extraction (MHMS) module is introduced to extract fine-grained information at various scales and increase the separability of facial features in high-dimensional space. Experimental results demonstrate that the LGAF network achieves state-of-the-art performance on four validation sets and outperforms existing methods on two additional datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to recognize low-quality face images by using both local and global features. This is important because sometimes parts of a face are missing or distorted, making it harder for computers to recognize the person. The authors suggest that both local and global features are needed to get good results. They also introduce a new module that helps extract more detailed information from faces. The experiments show that this approach works well and outperforms other methods. |
Keywords
» Artificial intelligence » Attention » Feature extraction
