Mining Multi-Modality Spatio-Temporal Cues for Video Important Person Identification 事件

PRODUCT_LAUNCH2026-05-28影响: MEDIUM

Mining Multi-Modality Spatio-Temporal Cues for Video Important Person Identification arXiv:2605.28604v1 Announce Type: new Abstract: Identifying key individuals in video scenes is essential for applications such as automated video editing and intelligent surveillance. Current methods primarily focus on static images and immediate visual cues, overlooking the rich spatio-temporal information in videos. This leads to the phenomenon of Temporal Importance Shift (TIS), wherein individuals deemed si