Summary of Accelerating Non-maximum Suppression: a Graph Theory Perspective, by King-siong Si et al.
Accelerating Non-Maximum Suppression: A Graph Theory Perspective
by King-Siong Si, Lu Sun, Weizhan Zhang, Tieliang Gong, Jiahao Wang, Jiang Liu, Hao Sun
First submitted to arxiv on: 30 Sep 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 This paper systematically analyzes non-maximum suppression (NMS) from a graph theory perspective for the first time, revealing its intrinsic structure. The authors propose two optimization methods, QSI-NMS and BOE-NMS, which respectively achieve 6.2x and 5.1x speedups over original NMS while maintaining mAP performance. Additionally, they introduce NMS-Bench, a comprehensive benchmark for assessing various NMS methods. The proposed methods are evaluated using the YOLOv8-N model on MS COCO 2017, achieving significant speed improvements with minimal impact on mAP. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper studies how to make object detection faster and better. It looks at something called non-maximum suppression (NMS) from a new perspective, which helps us understand how it works. The authors then come up with two ways to improve NMS: QSI-NMS and BOE-NMS. These methods are really fast, with QSI-NMS being 6.2 times faster than usual and BOE-NMS being 5.1 times faster. They also create a special tool called NMS-Bench that helps researchers test their ideas. The results show that these new methods can be very helpful. |
Keywords
» Artificial intelligence » Object detection » Optimization
