MotionStone: Decoupled Motion Intensity Modulation with Diffusion Transformer for Image-to-Video Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Shuwei, Gong, Biao, Chen, Xi, Zheng, Dandan, Tan, Shuai, Yang, Zizheng, Li, Yuyuan, He, Jingwen, Zheng, Kecheng, Chen, Jingdong, Yang, Ming, Zheng, Yinqiang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912147948175360
author Shi, Shuwei
Gong, Biao
Chen, Xi
Zheng, Dandan
Tan, Shuai
Yang, Zizheng
Li, Yuyuan
He, Jingwen
Zheng, Kecheng
Chen, Jingdong
Yang, Ming
Zheng, Yinqiang
author_facet Shi, Shuwei
Gong, Biao
Chen, Xi
Zheng, Dandan
Tan, Shuai
Yang, Zizheng
Li, Yuyuan
He, Jingwen
Zheng, Kecheng
Chen, Jingdong
Yang, Ming
Zheng, Yinqiang
contents The image-to-video (I2V) generation is conditioned on the static image, which has been enhanced recently by the motion intensity as an additional control signal. These motion-aware models are appealing to generate diverse motion patterns, yet there lacks a reliable motion estimator for training such models on large-scale video set in the wild. Traditional metrics, e.g., SSIM or optical flow, are hard to generalize to arbitrary videos, while, it is very tough for human annotators to label the abstract motion intensity neither. Furthermore, the motion intensity shall reveal both local object motion and global camera movement, which has not been studied before. This paper addresses the challenge with a new motion estimator, capable of measuring the decoupled motion intensities of objects and cameras in video. We leverage the contrastive learning on randomly paired videos and distinguish the video with greater motion intensity. Such a paradigm is friendly for annotation and easy to scale up to achieve stable performance on motion estimation. We then present a new I2V model, named MotionStone, developed with the decoupled motion estimator. Experimental results demonstrate the stability of the proposed motion estimator and the state-of-the-art performance of MotionStone on I2V generation. These advantages warrant the decoupled motion estimator to serve as a general plug-in enhancer for both data processing and video generation training.
format Preprint
id arxiv_https___arxiv_org_abs_2412_05848
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MotionStone: Decoupled Motion Intensity Modulation with Diffusion Transformer for Image-to-Video Generation
Shi, Shuwei
Gong, Biao
Chen, Xi
Zheng, Dandan
Tan, Shuai
Yang, Zizheng
Li, Yuyuan
He, Jingwen
Zheng, Kecheng
Chen, Jingdong
Yang, Ming
Zheng, Yinqiang
Computer Vision and Pattern Recognition
The image-to-video (I2V) generation is conditioned on the static image, which has been enhanced recently by the motion intensity as an additional control signal. These motion-aware models are appealing to generate diverse motion patterns, yet there lacks a reliable motion estimator for training such models on large-scale video set in the wild. Traditional metrics, e.g., SSIM or optical flow, are hard to generalize to arbitrary videos, while, it is very tough for human annotators to label the abstract motion intensity neither. Furthermore, the motion intensity shall reveal both local object motion and global camera movement, which has not been studied before. This paper addresses the challenge with a new motion estimator, capable of measuring the decoupled motion intensities of objects and cameras in video. We leverage the contrastive learning on randomly paired videos and distinguish the video with greater motion intensity. Such a paradigm is friendly for annotation and easy to scale up to achieve stable performance on motion estimation. We then present a new I2V model, named MotionStone, developed with the decoupled motion estimator. Experimental results demonstrate the stability of the proposed motion estimator and the state-of-the-art performance of MotionStone on I2V generation. These advantages warrant the decoupled motion estimator to serve as a general plug-in enhancer for both data processing and video generation training.
title MotionStone: Decoupled Motion Intensity Modulation with Diffusion Transformer for Image-to-Video Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.05848