Summary of Domain Similarity-perceived Label Assignment For Domain Generalized Underwater Object Detection, by Xisheng Li et al.
Domain Similarity-Perceived Label Assignment for Domain Generalized Underwater Object Detection
by Xisheng Li, Wei Li, Pinhao Song, Mingjun Zhang, Jie Zhou
First submitted to arxiv on: 20 Dec 2023
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 a new approach to tackle the issue of domain shift in underwater environments, where different layers and regions exhibit distinct characteristics. The Domain Adversarial Learning (DAL) strategy has been previously used to address this challenge, but it relies on manually assigned one-hot domain labels, which can lead to instability. To overcome this limitation, the authors introduce Domain Similarity-Perceived Label Assignment (DSP), a method that assigns domain labels based on an image’s similarity to specified domains. This approach achieves state-of-the-art results on the underwater cross-domain object detection benchmark S-UODAC2020 and also demonstrates effectiveness in the Cityscapes dataset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about how to make computers understand underwater environments better. When we take pictures or collect data from different places under water, it’s hard for machines to recognize what they’re seeing because each place has its own unique features. The authors came up with a new way to help machines learn about these differences and still be able to recognize things accurately. They call it Domain Similarity-Perceived Label Assignment (DSP) and it works really well in recognizing objects underwater. |
Keywords
» Artificial intelligence » Object detection » One hot
