CodecCap: High-Fidelity Codec-Inspired Residual Modeling for Dense Video Captioning 文章

ArXiv CS.CV2026-05-27NEWSen作者: Zihan Lin, Songhe Deng, Shuwei He, Danxiang Zhu, Dan Zhang, Yishu Lei, Xianlong Luo, Shikun Feng, Rui Liu
查看原文 →
关系图谱

CodecCap: High-Fidelity Codec-Inspired Residual Modeling for Dense Video Captioning · 相关技术

相关技术

ODEdivide-and-conquer partitioningParts-of-Speech (POS) tagsIDCHISGrouped Memorization EvaluationForDenseCamouflaged object detectionANNTable