Most segmentation models are trained on urban datasets (Cityscapes, ADE20K) and fail catastrophically in desert terrain — where there are no lane markings, few rigid objects, and extreme texture ...

Abstract: Video semantic segmentation aims to assign a semantic label to each pixel in a video by jointly exploiting spatial context and temporal coherence, and it is essential for applications such ...

Abstract: Light detection and ranging (LiDAR) point cloud semantic segmentation has emerged as a fundamental technique for enabling intelligent perception in autonomous driving, robotics, and ...

Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image ...

We’re introducing SAM 3 and SAM 3D, the newest additions to our Segment Anything Collection, which advance AI understanding of the visual world. SAM 3 enables detection and tracking of objects in ...