Abstract: Multiorgan segmentation in total-body positron emission tomography (PET) images is crucial for accurately locating abnormalities and assisting in the observation of corresponding metabolic ...

Abstract: Achieving effective and efficient segmentation of human body regions in distorted images is of practical significance. Current methods rely on transformers to extract discriminative features ...

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 ...

According to @AIatMeta, the Segment Anything Models (SAM) are being leveraged by the USRA and USGS to automate real-time river mapping, a critical bottleneck in flood monitoring and disaster response.