Summary of Physaug: a Physical-guided and Frequency-based Data Augmentation For Single-domain Generalized Object Detection, by Xiaoran Xu et al.
PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detection
by Xiaoran Xu, Jiangang Yang, Wenhui Shi, Siyuan Ding, Luqing Luo, Jian Liu
First submitted to arxiv on: 16 Dec 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 Single-Domain Generalized Object Detection (S-DGOD) paper proposes PhysAug, a novel data augmentation method that leverages physical model-based non-ideal imaging condition perturbations to enhance the adaptability of S-DGOD tasks. By simulating real-world variations in training data using atmospheric optics principles, PhysAug fosters domain-invariant representations and improves generalizability across various settings. The approach outperforms state-of-the-art methods on DWD and Cityscape-C datasets by 7.3% and 7.2%, respectively. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary PhysAug is a new way to make object detection models work better in different situations. Right now, models are only good at detecting objects in one specific setting. PhysAug helps the model learn how to detect objects in many different settings by adding fake imperfections to the training data. This makes the model more robust and able to generalize well across various scenarios. |
Keywords
» Artificial intelligence » Data augmentation » Object detection
