Summary of State Of the Art Applications Of Deep Learning Within Tracking and Detecting Marine Debris: a Survey, by Zoe Moorton et al.
State of the art applications of deep learning within tracking and detecting marine debris: A survey
by Zoe Moorton, Zeyneb Kurt, Wai Lok Woo
First submitted to arxiv on: 26 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); 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 abstract presents a comprehensive analysis of recent deep learning contributions to the marine litter problem, summarizing the findings of 28 significant research papers from the last five years. The study highlights that the YOLO family outperforms other object detection methods, but emphasizes the need for a comprehensive database of underwater debris for machine learning applications. Using a small curated dataset, the authors tested YOLOv5 on a binary classification task and found low accuracy and high false positive rates, underscoring the importance of a reliable database. The survey concludes with over 40 future research recommendations and open challenges. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Deep learning is being used to help solve the problem of marine litter. Scientists have been studying this issue for about 20 years, but most of the recent progress has happened in just the last five years. This paper looks at the 28 most important deep learning studies on marine debris from the last few years. It finds that a family of methods called YOLO is very good at detecting objects underwater. However, many experts agree that we need a big database of underwater trash to make machine learning work effectively. The authors test one type of YOLO method and find it’s not very accurate when working with the small dataset they used. They think this shows how important it is to have a reliable database. |
Keywords
* Artificial intelligence * Classification * Deep learning * Machine learning * Object detection * Yolo
