Separate Motion from Appearance: Customizing Motion via Customizing Text-to-Video Diffusion Models
- URL: http://arxiv.org/abs/2501.16714v1
- Date: Tue, 28 Jan 2025 05:40:20 GMT
- Title: Separate Motion from Appearance: Customizing Motion via Customizing Text-to-Video Diffusion Models
- Authors: Huijie Liu, Jingyun Wang, Shuai Ma, Jie Hu, Xiaoming Wei, Guoliang Kang,
- Abstract summary: Motion customization aims to adapt the diffusion model (DM) to generate videos with the motion specified by a set of video clips with the same motion concept.
This paper proposes two novel strategies to enhance motion-appearance separation, including temporal attention purification (TAP) and appearance highway (AH)
- Score: 18.41701130228042
- License:
- Abstract: Motion customization aims to adapt the diffusion model (DM) to generate videos with the motion specified by a set of video clips with the same motion concept. To realize this goal, the adaptation of DM should be possible to model the specified motion concept, without compromising the ability to generate diverse appearances. Thus, the key to solving this problem lies in how to separate the motion concept from the appearance in the adaptation process of DM. Typical previous works explore different ways to represent and insert a motion concept into large-scale pretrained text-to-video diffusion models, e.g., learning a motion LoRA, using latent noise residuals, etc. While those methods can encode the motion concept, they also inevitably encode the appearance in the reference videos, resulting in weakened appearance generation capability. In this paper, we follow the typical way to learn a motion LoRA to encode the motion concept, but propose two novel strategies to enhance motion-appearance separation, including temporal attention purification (TAP) and appearance highway (AH). Specifically, we assume that in the temporal attention module, the pretrained Value embeddings are sufficient to serve as basic components needed by producing a new motion. Thus, in TAP, we choose only to reshape the temporal attention with motion LoRAs so that Value embeddings can be reorganized to produce a new motion. Further, in AH, we alter the starting point of each skip connection in U-Net from the output of each temporal attention module to the output of each spatial attention module. Extensive experiments demonstrate that compared to previous works, our method can generate videos with appearance more aligned with the text descriptions and motion more consistent with the reference videos.
Related papers
- MotionMatcher: Motion Customization of Text-to-Video Diffusion Models via Motion Feature Matching [27.28898943916193]
Text-to-video (T2V) diffusion models have promising capabilities in synthesizing realistic videos from input text prompts.
In this work, we tackle the motion customization problem, where a reference video is provided as motion guidance.
We propose MotionMatcher, a motion customization framework that fine-tunes the pre-trained T2V diffusion model at the feature level.
arXiv Detail & Related papers (2025-02-18T19:12:51Z) - MoTrans: Customized Motion Transfer with Text-driven Video Diffusion Models [59.10171699717122]
MoTrans is a customized motion transfer method enabling video generation of similar motion in new context.
multimodal representations from recaptioned prompt and video frames promote the modeling of appearance.
Our method effectively learns specific motion pattern from singular or multiple reference videos.
arXiv Detail & Related papers (2024-12-02T10:07:59Z) - CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities [56.5742116979914]
CustomCrafter preserves the model's motion generation and conceptual combination abilities without additional video and fine-tuning to recovery.
For motion generation, we observed that VDMs tend to restore the motion of video in the early stage of denoising.
In the later stage of denoising, we restore this module to repair the appearance details of the specified subject.
arXiv Detail & Related papers (2024-08-23T17:26:06Z) - Motion Inversion for Video Customization [31.607669029754874]
We present a novel approach for motion in generation, addressing the widespread gap in the exploration of motion representation within video models.
We introduce Motion Embeddings, a set of explicit, temporally coherent embeddings derived from given video.
Our contributions include a tailored motion embedding for customization tasks and a demonstration of the practical advantages and effectiveness of our method.
arXiv Detail & Related papers (2024-03-29T14:14:22Z) - Animate Your Motion: Turning Still Images into Dynamic Videos [58.63109848837741]
We introduce Scene and Motion Conditional Diffusion (SMCD), a novel methodology for managing multimodal inputs.
SMCD incorporates a recognized motion conditioning module and investigates various approaches to integrate scene conditions.
Our design significantly enhances video quality, motion precision, and semantic coherence.
arXiv Detail & Related papers (2024-03-15T10:36:24Z) - MotionCrafter: One-Shot Motion Customization of Diffusion Models [66.44642854791807]
We introduce MotionCrafter, a one-shot instance-guided motion customization method.
MotionCrafter employs a parallel spatial-temporal architecture that injects the reference motion into the temporal component of the base model.
During training, a frozen base model provides appearance normalization, effectively separating appearance from motion.
arXiv Detail & Related papers (2023-12-08T16:31:04Z) - NewMove: Customizing text-to-video models with novel motions [74.9442859239997]
We introduce an approach for augmenting text-to-video generation models with customized motions.
By leveraging a few video samples demonstrating specific movements as input, our method learns and generalizes the input motion patterns for diverse, text-specified scenarios.
arXiv Detail & Related papers (2023-12-07T18:59:03Z) - VMC: Video Motion Customization using Temporal Attention Adaption for
Text-to-Video Diffusion Models [58.93124686141781]
Video Motion Customization (VMC) is a novel one-shot tuning approach crafted to adapt temporal attention layers within video diffusion models.
Our approach introduces a novel motion distillation objective using residual vectors between consecutive frames as a motion reference.
We validate our method against state-of-the-art video generative models across diverse real-world motions and contexts.
arXiv Detail & Related papers (2023-12-01T06:50:11Z) - MotionDirector: Motion Customization of Text-to-Video Diffusion Models [24.282240656366714]
Motion Customization aims to adapt existing text-to-video diffusion models to generate videos with customized motion.
We propose MotionDirector, with a dual-path LoRAs architecture to decouple the learning of appearance and motion.
Our method also supports various downstream applications, such as the mixing of different videos with their appearance and motion respectively, and animating a single image with customized motions.
arXiv Detail & Related papers (2023-10-12T16:26:18Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.