Saved in:
Bibliographic Details
Main Authors: Sun, Mingzhen, Wang, Weining, Zhu, Xinxin, Liu, Jing
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.01718
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914962785435648
author Sun, Mingzhen
Wang, Weining
Zhu, Xinxin
Liu, Jing
author_facet Sun, Mingzhen
Wang, Weining
Zhu, Xinxin
Liu, Jing
contents Since videos record objects moving coherently, adjacent video frames have commonness (similar object appearances) and uniqueness (slightly changed postures). To prevent redundant modeling of common video signals, we propose a novel diffusion-based framework, named COMUNI, which decomposes the COMmon and UNIque video signals to enable efficient video generation. Our approach separates the decomposition of video signals from the task of video generation, thus reducing the computation complexity of generative models. In particular, we introduce CU-VAE to decompose video signals and encode them into latent features. To train CU-VAE in a self-supervised manner, we employ a cascading merge module to reconstitute video signals and a time-agnostic video decoder to reconstruct video frames. Then we propose CU-LDM to model latent features for video generation, which adopts two specific diffusion streams to simultaneously model the common and unique latent features. We further utilize additional joint modules for cross modeling of the common and unique latent features, and a novel position embedding method to ensure the content consistency and motion coherence of generated videos. The position embedding method incorporates spatial and temporal absolute position information into the joint modules. Extensive experiments demonstrate the necessity of decomposing common and unique video signals for video generation and the effectiveness and efficiency of our proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01718
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle COMUNI: Decomposing Common and Unique Video Signals for Diffusion-based Video Generation
Sun, Mingzhen
Wang, Weining
Zhu, Xinxin
Liu, Jing
Computer Vision and Pattern Recognition
Since videos record objects moving coherently, adjacent video frames have commonness (similar object appearances) and uniqueness (slightly changed postures). To prevent redundant modeling of common video signals, we propose a novel diffusion-based framework, named COMUNI, which decomposes the COMmon and UNIque video signals to enable efficient video generation. Our approach separates the decomposition of video signals from the task of video generation, thus reducing the computation complexity of generative models. In particular, we introduce CU-VAE to decompose video signals and encode them into latent features. To train CU-VAE in a self-supervised manner, we employ a cascading merge module to reconstitute video signals and a time-agnostic video decoder to reconstruct video frames. Then we propose CU-LDM to model latent features for video generation, which adopts two specific diffusion streams to simultaneously model the common and unique latent features. We further utilize additional joint modules for cross modeling of the common and unique latent features, and a novel position embedding method to ensure the content consistency and motion coherence of generated videos. The position embedding method incorporates spatial and temporal absolute position information into the joint modules. Extensive experiments demonstrate the necessity of decomposing common and unique video signals for video generation and the effectiveness and efficiency of our proposed method.
title COMUNI: Decomposing Common and Unique Video Signals for Diffusion-based Video Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.01718