Medical images are the standard approach for the analysis and diagnosis of critical issues of diseases. To minimize the time-consuming inspection and evaluation process of the medical images from ...

This paper considers how to separate text and/or graphics from smooth background in screen content and mixed document images and proposes two approaches to perform this segmentation task. The proposed ...

Recent advances in AI have revolutionized Earth observation, yet most remote sensing tasks still rely on specialized models with fragmented interfaces. To address this, we present RemoteSAM, a vision ...

Then, a VLM for farmland segmentation (FSVLM) was designed by combining a semantic segmentation model with a multimodal large language model (LLM). Comparative experiments demonstrated that the ...

• GA-TongueNet is proposed for tongue image semantic segmentation, effectively capturing long-range dependencies, high-level semantic information, and low-level detail information. The network ...