Summary of Semi-supervised Image Captioning Considering Wasserstein Graph Matching, by Yang Yang
Semi-Supervised Image Captioning Considering Wasserstein Graph Matching
by Yang Yang
First submitted to arxiv on: 26 Mar 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 Semi-Supervised Image Captioning method, SSIC-WGM, tackles the challenge of limited described images in real-world applications by adopting Wasserstein Graph Matching. It constrains generated sentences from inter-modal consistency, ensuring similarity between region embeddings of raw image scene graphs and generated sentence scene graphs, as well as intra-modal consistency through data augmentation techniques. By combining cross-modal pseudo supervision and structure invariant measure, SSIC-WGM efficiently utilizes undescribed images to learn a more reasonable mapping function. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine having an AI that can automatically describe pictures! This technology is called image captioning, but it’s tricky because we usually don’t have many descriptions for the same number of pictures. The researchers developed a new way to make this work with limited information. They used something called Wasserstein Graph Matching to connect the pictures and words together. This helps the AI learn how to describe pictures even when there aren’t many examples. |
Keywords
* Artificial intelligence * Data augmentation * Image captioning * Semi supervised
