Summary of Open-vocabulary X-ray Prohibited Item Detection Via Fine-tuning Clip, by Shuyang Lin et al.
Open-Vocabulary X-ray Prohibited Item Detection via Fine-tuning CLIP
by Shuyang Lin, Tong Jia, Hao Wang, Bowen Ma, Mingyuan Li, Dongyue Chen
First submitted to arxiv on: 16 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
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 approach addresses the challenges of open-set X-ray prohibited item detection by introducing a distillation-based open-vocabulary object detection (OVOD) task, extending CLIP to learn visual representations in the specific X-ray domain. The OVXD model incorporates an X-ray feature adapter, which bridges the domain gap and promotes alignment between X-ray images and textual concepts. This approach outperforms previous best results by 15.2 AP50 on PIXray and PIDray datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper introduces a new way to detect prohibited items in X-rays that can recognize unknown categories. Currently, methods only work well for known categories and require lots of training data. The researchers use a special type of AI model called CLIP and adapt it to learn about X-ray images. They also propose a new module called the X-ray feature adapter, which helps the model understand the specific features of X-ray images better. This approach is tested on two datasets and performs well, achieving better results than previous methods. |
Keywords
» Artificial intelligence » Alignment » Distillation » Object detection
