Summary of Towards Temporal Change Explanations From Bi-temporal Satellite Images, by Ryo Tsujimoto et al.
Towards Temporal Change Explanations from Bi-Temporal Satellite Images
by Ryo Tsujimoto, Hiroki Ouchi, Hidetaka Kamigaito, Taro Watanabe
First submitted to arxiv on: 27 Jun 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 This paper explores the ability of Large-scale Vision-Language Models (LVLMs) to explain temporal changes between satellite images taken at different times. Manual dataset construction for this task is costly and time-consuming, making human-AI collaboration a promising approach. The authors investigate how LVLMs can be used to capture temporal changes between satellite images, proposing three prompting methods to deal with the unique challenge of processing pairs of images as input. Through human evaluation, the authors found that their step-by-step reasoning-based prompting method is effective in leveraging LVLMs for this task. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine trying to understand what’s changed in a city over time using only old and new satellite photos. This is a big challenge because we need to compare two images taken at different times. Right now, it takes a lot of work to create the data needed to do this. To make things easier, humans can work together with computers to help solve this problem. In this paper, researchers try to figure out if powerful computer models called Large-scale Vision-Language Models (LVLMs) can be used to understand what’s changed in satellite images over time. They came up with new ways to ask these computer models questions and found that one of their methods works really well. |
Keywords
» Artificial intelligence » Prompting
