GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video Retrieval

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yuting, Wang, Jinpeng, Chen, Bin, Zeng, Ziyun, Xia, Shu-Tao
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914627879698432
author Wang, Yuting
Wang, Jinpeng
Chen, Bin
Zeng, Ziyun
Xia, Shu-Tao
author_facet Wang, Yuting
Wang, Jinpeng
Chen, Bin
Zeng, Ziyun
Xia, Shu-Tao
contents Given a text query, partially relevant video retrieval (PRVR) seeks to find untrimmed videos containing pertinent moments in a database. For PRVR, clip modeling is essential to capture the partial relationship between texts and videos. Current PRVR methods adopt scanning-based clip construction to achieve explicit clip modeling, which is information-redundant and requires a large storage overhead. To solve the efficiency problem of PRVR methods, this paper proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly. During frame interactions, we incorporate Gaussian-Mixture-Model constraints to focus each frame on its adjacent frames instead of the whole video. Then generated representations will contain multi-scale clip information, achieving implicit clip modeling. In addition, PRVR methods ignore semantic differences between text queries relevant to the same video, leading to a sparse embedding space. We propose a query diverse loss to distinguish these text queries, making the embedding space more intensive and contain more semantic information. Extensive experiments on three large-scale video datasets (i.e., TVR, ActivityNet Captions, and Charades-STA) demonstrate the superiority and efficiency of GMMFormer. Code is available at \url{https://github.com/huangmozhi9527/GMMFormer}.
format Preprint
id arxiv_https___arxiv_org_abs_2310_05195
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video Retrieval
Wang, Yuting
Wang, Jinpeng
Chen, Bin
Zeng, Ziyun
Xia, Shu-Tao
Computer Vision and Pattern Recognition
Artificial Intelligence
Information Retrieval
Multimedia
Given a text query, partially relevant video retrieval (PRVR) seeks to find untrimmed videos containing pertinent moments in a database. For PRVR, clip modeling is essential to capture the partial relationship between texts and videos. Current PRVR methods adopt scanning-based clip construction to achieve explicit clip modeling, which is information-redundant and requires a large storage overhead. To solve the efficiency problem of PRVR methods, this paper proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly. During frame interactions, we incorporate Gaussian-Mixture-Model constraints to focus each frame on its adjacent frames instead of the whole video. Then generated representations will contain multi-scale clip information, achieving implicit clip modeling. In addition, PRVR methods ignore semantic differences between text queries relevant to the same video, leading to a sparse embedding space. We propose a query diverse loss to distinguish these text queries, making the embedding space more intensive and contain more semantic information. Extensive experiments on three large-scale video datasets (i.e., TVR, ActivityNet Captions, and Charades-STA) demonstrate the superiority and efficiency of GMMFormer. Code is available at \url{https://github.com/huangmozhi9527/GMMFormer}.
title GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video Retrieval
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Information Retrieval
Multimedia
url https://arxiv.org/abs/2310.05195