Summary of Urban Waterlogging Detection: a Challenging Benchmark and Large-small Model Co-adapter, by Suqi Song et al.
Urban Waterlogging Detection: A Challenging Benchmark and Large-Small Model Co-Adapter
by Suqi Song, Chenxu Zhang, Peng Zhang, Pengkun Li, Fenglong Song, Lei Zhang
First submitted to arxiv on: 11 Jul 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 proposed Large-Small Model co-adapter paradigm (LSM-adapter) combines the strengths of large and small models to improve urban waterlogging detection. The authors create a challenging Urban Waterlogging Benchmark (UW-Bench) to test their approach, which includes a Triple-S Prompt Adapter module and Dynamic Prompt Combiner for adapting mask decoders. A Histogram Equalization Adap-ter module is also designed to adapt image encoders. The results demonstrate the effectiveness of the developed benchmark and algorithm. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Urban waterlogging is a big problem that affects public safety and infrastructure. Right now, there are not many good ways to detect it because traditional methods require a lot of maintenance. Newer approaches using deep learning and camera images can work well when they have plenty of data and good conditions. But in real-world situations, things are different. In this paper, the authors create a new benchmark called Urban Waterlogging Benchmark (UWB) that tests how well models do under tough conditions. They also propose an approach called LSM-adapter that uses both large and small models to improve detection. |
Keywords
» Artificial intelligence » Deep learning » Mask » Prompt
