Summary of Toward a More Complete Omr Solution, by Guang Yang (1) et al.
Toward a More Complete OMR Solution
by Guang Yang, Muru Zhang, Lin Qiu, Yanming Wan, Noah A. Smith
First submitted to arxiv on: 31 Aug 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 paper proposes an innovative approach to optical music recognition (OMR), a technology that converts music notation into digital formats. The method involves a multi-stage pipeline, where the system first detects visual music notation elements in an image using object detection and then assembles them into a music notation. Previous work on OMR has unrealistically assumed perfect object detection, but this study tackles the challenge by considering both stages together. The researchers introduce a music object detector based on YOLOv8, which improves detection performance, and develop a supervised training pipeline that completes the notation assembly stage based on detection output. The results show that this model outperforms existing models trained on perfect detection output, highlighting the importance of considering the detection and assembly stages in a more holistic way. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about making computers understand music written down. Right now, we have ways to read printed music, but they’re not perfect. The researchers wanted to make it better by combining two tasks: recognizing what’s on the page (like notes, staff lines) and putting those pieces together into a complete piece of music. They used a special computer program called YOLOv8 to help identify the individual musical elements, then another part to put them together into a correct score. This approach worked better than previous methods that assumed everything was perfect, which isn’t always the case. This new way could lead to more accurate music recognition in the future. |
Keywords
* Artificial intelligence * Object detection * Supervised
