dastovek/samurai

The SAMURAI project is a zero-shot visual tracking method that adapts the Segment Anything Model (SAM) for tracking objects in videos with motion-aware memory. It is designed for researchers and developers in the field of computer vision, particularly those interested in object tracking and segmentation. The project provides a pre-trained model and code for inference on various video tracking benchmarks, making it a useful resource for those looking to explore state-of-the-art tracking methods.