Summary of Fine-tuning Vision-language Model For Automated Engineering Drawing Information Extraction, by Muhammad Tayyab Khan et al.
Fine-Tuning Vision-Language Model for Automated Engineering Drawing Information Extraction
by Muhammad Tayyab Khan, Lequn Chen, Ye Han Ng, Wenhe Feng, Nicholas Yew Jin Tan, Seung Ki Moon
First submitted to arxiv on: 6 Nov 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 study proposes an automated method for extracting Geometric Dimensioning and Tolerancing (GD&T) information from 2D engineering drawings. The approach involves fine-tuning the open-source vision-language model Florence-2, which is trained on a dataset of 400 drawings with ground truth annotations provided by domain experts. The model is assessed using precision, recall, F1-score, and hallucination metrics, achieving significant improvements over state-of-the-art closed-source models. The results highlight the effectiveness of fine-tuning smaller, open-source vision-language models like Florence-2 for practical and efficient automated GD&T extraction. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps computers understand engineering drawings better. It’s hard to get the right information from these drawings now, so people are trying different ways to make it easier. One way is to use a special kind of computer program that can read and understand pictures. The researchers used a small open-source program called Florence-2 and trained it on lots of examples. They compared it to bigger, more powerful programs and found that Florence-2 was much better at getting the right information. This could help people make things faster and cheaper by letting computers do some of the work. |
Keywords
» Artificial intelligence » F1 score » Fine tuning » Hallucination » Language model » Precision » Recall
