دورية أكاديمية
Practical Techniques for Vision-Language Segmentation Model in Remote Sensing
العنوان: | Practical Techniques for Vision-Language Segmentation Model in Remote Sensing |
---|---|
المؤلفون: | Y. Lin, K. Suzuki, S. Sogo |
المصدر: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XLVIII-2-2024, Pp 203-210 (2024) |
بيانات النشر: | Copernicus Publications, 2024. |
سنة النشر: | 2024 |
المجموعة: | LCC:Technology LCC:Engineering (General). Civil engineering (General) LCC:Applied optics. Photonics |
مصطلحات موضوعية: | Technology, Engineering (General). Civil engineering (General), TA1-2040, Applied optics. Photonics, TA1501-1820 |
الوصف: | Traditional semantic segmentation models often struggle with poor generalizability in zero-shot scenarios such as recognizing attributes unseen in the training labels. On the other hands, language-vision models (VLMs) have shown promise in improving performance on zero-shot tasks by leveraging semantic information from textual inputs and fusing this information with visual features. However, existing VLM-based methods do not perform as effectively on remote sensing data due to the lack of such data in their training datasets. In this paper, we introduce a two-stage fine-tuning approach for a VLM-based segmentation model using a large remote sensing image-caption dataset, which we created using an existing image-caption model. Additionally, we propose a modified decoder and a visual prompt technique using a saliency map to enhance segmentation results. Through these methods, we achieve superior segmentation performance on remote sensing data, demonstrating the effectiveness of our approach. |
نوع الوثيقة: | article |
وصف الملف: | electronic resource |
اللغة: | English |
تدمد: | 1682-1750 2194-9034 |
Relation: | https://isprs-archives.copernicus.org/articles/XLVIII-2-2024/203/2024/isprs-archives-XLVIII-2-2024-203-2024.pdf; https://doaj.org/toc/1682-1750; https://doaj.org/toc/2194-9034 |
DOI: | 10.5194/isprs-archives-XLVIII-2-2024-203-2024 |
URL الوصول: | https://doaj.org/article/7e8446aedbdc4aa5b2803db9eb0912bc |
رقم الأكسشن: | edsdoj.7e8446aedbdc4aa5b2803db9eb0912bc |
قاعدة البيانات: | Directory of Open Access Journals |
تدمد: | 16821750 21949034 |
---|---|
DOI: | 10.5194/isprs-archives-XLVIII-2-2024-203-2024 |