Abstract: Current prevailing Video Object Segmentation (VOS) methods follow the pipeline of extraction-then-matching, which first extracts features on current and reference frames independently, and ...

Abstract: Convolutional neural networks (CNNs) are frequently used to analyze remote sensing images and achieve impressive progress. Limited by the receptive field size of CNNs, small objects tended ...

ObjectClear is an object removal model that can jointly eliminate the target object and its associated effects leveraging Object-Effect Attention, while preserving background consistency. OBER (OBject ...

The repo is the official implement of "Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement". The repo is also the official training code of "VisionReasoner: Unified Visual ...