MiVE: Multiscale Vision-language features for reference-guided video Editing 文章

ArXiv CS.CV2026-05-27NEWSen作者: Tong Wang, Meng Zou, Chengjing Wu, Xiaochao Qu, Luoqi Liu, Xiaolin Hu, Ting Liu
查看原文 →
关系图谱

MiVE: Multiscale Vision-language features for reference-guided video Editing · 相关技术

相关技术

ginORMODEPLA远程代码执行（RCE）divide-and-conquer partitioningUnifiedUCTSSESPASEMReferring expression comprehension (REC)Parts-of-Speech (POS) tagsNPUMITVLMHISGranular Alignment ParadigmForEffort Metric AttentionDiTCamouflaged object detectionCISANN