Summary of Msmsfnet: a Multi-stream and Multi-scale Fusion Net For Edge Detection, by Chenguang Liu et al.
Msmsfnet: a multi-stream and multi-scale fusion net for edge detection
by Chenguang Liu, Chisheng Wang, Feifei Dong, Xiayang Xiao, Xin Su, Chuanhua Zhu, Dejin Zhang, Qingquan Li
First submitted to arxiv on: 7 Apr 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 A novel deep learning-based approach to edge detection in computer vision has been proposed, departing from traditional methods that rely heavily on pre-trained weights from the ImageNet dataset. The existing state-of-the-art algorithms achieve excellent performance in publicly available datasets but are limited by their dependence on these pre-trained weights. This paper investigates the performance of these methods when trained from scratch and proposes a new architecture, msmsfnet, which outperforms current deep learning-based edge detectors in three public datasets. Moreover, the authors demonstrate the effectiveness of their approach for edge detection in Synthetic Aperture Radar (SAR) images, highlighting its potential for real-world applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers has developed a new way to detect edges in computer vision. They wanted to create a model that doesn’t rely on pre-trained weights from a big dataset called ImageNet. Instead, they started with a blank slate and trained their model from scratch. This allowed them to compare different approaches fairly, without any advantages given to some models. The team also designed a new network architecture, msmsfnet, which did very well in detecting edges in three real-world datasets. What’s more, their approach worked even better when they used it with Synthetic Aperture Radar (SAR) images, which are different from the usual types of images. |
Keywords
» Artificial intelligence » Deep learning
