Level-wise aligned dual networks for text–video retrieval

Abstract The vast amount of videos on the Internet makes efficient and accurate text–video retrieval tasks increasingly important. The current methods leverage a high-dimensional space to align video and text for these tasks. However, a high-dimensional space cannot fully use different levels of inf...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Qiubin Lin, Wenming Cao, Zhiquan He
Format:	Artigo
Jezik:	Inglês
Izdano:	SpringerOpen 2022-07-01
Serija:	EURASIP Journal on Advances in Signal Processing
Teme:	Text–video retrieval Level-wise aligned mechanism Semantic space Latent space
Online dostop:	https://doi.org/10.1186/s13634-022-00887-y
Oznake:	Označite Brez oznak, prvi označite!