Towards Automatic Video Annotation with ASH: Zero-Shot Open-Vocabulary Multi-Object Tracking and Segmentation figure
AlphaXiv 中文概览(可滚动查看)