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

TL;DR

MotionStone introduces a novel decoupled motion estimator for image-to-video generation, enabling more accurate control of local and global motions, and achieves state-of-the-art results with scalable, contrastive learning-based training.

Contribution

The paper proposes a new decoupled motion estimator using contrastive learning, improving motion measurement for I2V generation and enhancing model stability and performance.

Findings

The motion estimator accurately measures object and camera motion intensities.

MotionStone achieves state-of-the-art results on image-to-video generation tasks.

The decoupled estimator is scalable and serves as a versatile plug-in for various video processing applications.

Abstract

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.