Abstract: In the realm of medical image segmentation, the U-shaped structure comprising an encoder and a decoder has been demonstrated to be an extremely effective approach. They can be classified ...

We present SegLLM, a novel multi-round interactive segmentation model that leverages conversational memory of both visual and textual outputs to reason over previously segmented objects and past ...

Researchers at Tsinghua University developed PriorFusion, a unified framework that integrates semantic, geometric, and ...