Excite, Attend and Segment (EASe): Domain-Agnostic Fine-Grained Mask Discovery with Feature Calibration and Self-Supervised Upsampling

ArXi:2604.00276v1 Announce Type: new Unsupervised segmentation approaches have increasingly leveraged foundation models (FM) to improve salient object discovery. However, these methods often falter in scenes with complex, multi-component morphologies, where fine-grained structural detail is indispensable. Many state-of-the-art unsupervised segmentation pipelines rely on mask discovery approaches that utilize coarse, patch-level representations. These coarse representations inherently suppress the fine-grained detail required to resolve such complex morphologies.