SweetTok: Semantic-Aware Spatial-Temporal Tokenizer for Compact Video Discretization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tan, Zhentao, Xue, Ben, Jia, Jian, Wang, Junhao, Ye, Wencai, Shi, Shaoyun, Sun, Mingjie, Wu, Wenjin, Chen, Quan, Jiang, Peng
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917950775099392
author Tan, Zhentao
Xue, Ben
Jia, Jian
Wang, Junhao
Ye, Wencai
Shi, Shaoyun
Sun, Mingjie
Wu, Wenjin
Chen, Quan
Jiang, Peng
author_facet Tan, Zhentao
Xue, Ben
Jia, Jian
Wang, Junhao
Ye, Wencai
Shi, Shaoyun
Sun, Mingjie
Wu, Wenjin
Chen, Quan
Jiang, Peng
contents This paper presents the \textbf{S}emantic-a\textbf{W}ar\textbf{E} spatial-t\textbf{E}mporal \textbf{T}okenizer (SweetTok), a novel video tokenizer to overcome the limitations in current video tokenization methods for compacted yet effective discretization. Unlike previous approaches that process flattened local visual patches via direct discretization or adaptive query tokenization, SweetTok proposes a decoupling framework, compressing visual inputs through distinct spatial and temporal queries via \textbf{D}ecoupled \textbf{Q}uery \textbf{A}uto\textbf{E}ncoder (DQAE). This design allows SweetTok to efficiently compress video token count while achieving superior fidelity by capturing essential information across spatial and temporal dimensions. Furthermore, we design a \textbf{M}otion-enhanced \textbf{L}anguage \textbf{C}odebook (MLC) tailored for spatial and temporal compression to address the differences in semantic representation between appearance and motion information. SweetTok significantly improves video reconstruction results by \textbf{42.8\%} w.r.t rFVD on UCF-101 dataset. With a better token compression strategy, it also boosts downstream video generation results by \textbf{15.1\%} w.r.t gFVD. Additionally, the compressed decoupled tokens are imbued with semantic information, enabling few-shot recognition capabilities powered by LLMs in downstream applications.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10443
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SweetTok: Semantic-Aware Spatial-Temporal Tokenizer for Compact Video Discretization
Tan, Zhentao
Xue, Ben
Jia, Jian
Wang, Junhao
Ye, Wencai
Shi, Shaoyun
Sun, Mingjie
Wu, Wenjin
Chen, Quan
Jiang, Peng
Computer Vision and Pattern Recognition
Artificial Intelligence
This paper presents the \textbf{S}emantic-a\textbf{W}ar\textbf{E} spatial-t\textbf{E}mporal \textbf{T}okenizer (SweetTok), a novel video tokenizer to overcome the limitations in current video tokenization methods for compacted yet effective discretization. Unlike previous approaches that process flattened local visual patches via direct discretization or adaptive query tokenization, SweetTok proposes a decoupling framework, compressing visual inputs through distinct spatial and temporal queries via \textbf{D}ecoupled \textbf{Q}uery \textbf{A}uto\textbf{E}ncoder (DQAE). This design allows SweetTok to efficiently compress video token count while achieving superior fidelity by capturing essential information across spatial and temporal dimensions. Furthermore, we design a \textbf{M}otion-enhanced \textbf{L}anguage \textbf{C}odebook (MLC) tailored for spatial and temporal compression to address the differences in semantic representation between appearance and motion information. SweetTok significantly improves video reconstruction results by \textbf{42.8\%} w.r.t rFVD on UCF-101 dataset. With a better token compression strategy, it also boosts downstream video generation results by \textbf{15.1\%} w.r.t gFVD. Additionally, the compressed decoupled tokens are imbued with semantic information, enabling few-shot recognition capabilities powered by LLMs in downstream applications.
title SweetTok: Semantic-Aware Spatial-Temporal Tokenizer for Compact Video Discretization
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2412.10443