CaptionFormer: Unified Segmentation, Tracking, and Captioning for Spatio-Temporal Objects 文章

ArXiv CS.CV2026-06-02NEWSen作者: Gabriel Fiastre, Antoine Yang, Cordelia Schmid
查看原文 →
关系图谱

CaptionFormer: Unified Segmentation, Tracking, and Captioning for Spatio-Temporal Objects · 相关技术

相关技术

LVLMsDense Video Object CaptioningDVOC