Few-Shot-Based Modular Image-to-Video Adapter for Diffusion Models

• URL: http://arxiv.org/abs/2512.20000v2
• Date: Wed, 31 Dec 2025 02:39:39 GMT
• Title: Few-Shot-Based Modular Image-to-Video Adapter for Diffusion Models
• Abstract summary: Diffusion models (DMs) have recently achieved impressive photorealism in image and video generation.<n>DMs struggle to generalize to novel motion patterns not present in the training set.<n>We propose Modular Image-to-Video Adapter (MIVA), a lightweight sub-network attachable to a pre-trained DM.
• Score: 16.381660228819936
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Diffusion models (DMs) have recently achieved impressive photorealism in image and video generation. However, their application to image animation remains limited, even when trained on large-scale datasets. Two primary challenges contribute to this: the high dimensionality of video signals leads to a scarcity of training data, causing DMs to favor memorization over prompt compliance when generating motion; moreover, DMs struggle to generalize to novel motion patterns not present in the training set, and fine-tuning them to learn such patterns, especially using limited training data, is still under-explored. To address these limitations, we propose Modular Image-to-Video Adapter (MIVA), a lightweight sub-network attachable to a pre-trained DM, each designed to capture a single motion pattern and scalable via parallelization. MIVAs can be efficiently trained on approximately ten samples using a single consumer-grade GPU. At inference time, users can specify motion by selecting one or multiple MIVAs, eliminating the need for prompt engineering. Extensive experiments demonstrate that MIVA enables more precise motion control while maintaining, or even surpassing, the generation quality of models trained on significantly larger datasets.

Related papers

• MMPhysVideo: Scaling Physical Plausibility in Video Generation via Joint Multimodal Modeling [55.72785604682579]
  MMPhysVideo is a framework to scale physical plausibility in video generation in joint multimodal modeling.<n>MMPhysVideo consistently improves physical plausibility and visual quality over advanced models across various benchmarks.
  arXiv  Detail & Related papers  (2026-04-03T07:32:24Z)
• Moaw: Unleashing Motion Awareness for Video Diffusion Models [71.34328578845721]
  Moaw is a framework that unleashes motion awareness for video diffusion models.<n>We train a diffusion model for motion perception, shifting its modality from image-to-video generation to video-to-dense-tracking.<n>We then construct a motion-labeled dataset to identify features that encode the strongest motion information, and inject them into a structurally identical video generation model.
  arXiv  Detail & Related papers  (2026-01-19T06:45:46Z)
• MotionDiff: Training-free Zero-shot Interactive Motion Editing via Flow-assisted Multi-view Diffusion [20.142107033583027]
  MotionDiff is a training-free zero-shot diffusion method that leverages optical flow for complex multi-view motion editing.<n>It outperforms other physics-based generative motion editing methods in achieving high-quality multi-view consistent motion results.<n>MotionDiff does not require retraining, enabling users to conveniently adapt it for various down-stream tasks.
  arXiv  Detail & Related papers  (2025-03-22T08:32:56Z)
• 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.<n> multimodal representations from recaptioned prompt and video frames promote the modeling of appearance.<n>Our method effectively learns specific motion pattern from singular or multiple reference videos.
  arXiv  Detail & Related papers  (2024-12-02T10:07:59Z)
• AM Flow: Adapters for Temporal Processing in Action Recognition [6.67921694218089]
  "textitAttention Map (AM) Flow" is a method for identifying pixels relevant to motion in each input video frame. AM flow allows the separation spatial and temporal processing, while providing improved results over combined-temporal processing. We present experiments on Kinetics-400, Something-Something v2, and Toyota Smarthome datasets, showcasing state-of-the-art or comparable results.
  arXiv  Detail & Related papers  (2024-11-04T13:07:22Z)
• 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.<n>For motion generation, we observed that VDMs tend to restore the motion of video in the early stage of denoising.<n>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)
• AID: Adapting Image2Video Diffusion Models for Instruction-guided Video Prediction [88.70116693750452]
  Text-guided video prediction (TVP) involves predicting the motion of future frames from the initial frame according to an instruction. Previous TVP methods make significant breakthroughs by adapting Stable Diffusion for this task. We introduce the Multi-Modal Large Language Model (MLLM) to predict future video states based on initial frames and text instructions.
  arXiv  Detail & Related papers  (2024-06-10T17:02:08Z)
• Pix2Gif: Motion-Guided Diffusion for GIF Generation [70.64240654310754]
  We present Pix2Gif, a motion-guided diffusion model for image-to-GIF (video) generation. We propose a new motion-guided warping module to spatially transform the features of the source image conditioned on the two types of prompts. In preparation for the model training, we meticulously curated data by extracting coherent image frames from the TGIF video-caption dataset.
  arXiv  Detail & Related papers  (2024-03-07T16:18:28Z)
• Mug-STAN: Adapting Image-Language Pretrained Models for General Video Understanding [47.97650346560239]
  We propose Spatial-Temporal Auxiliary Network with Mutual-guided alignment module (Mug-STAN) to extend image-text model to diverse video tasks and video-text data. Mug-STAN significantly improves adaptation of language-image pretrained models such as CLIP and CoCa at both video-text post-pretraining and finetuning stages.
  arXiv  Detail & Related papers  (2023-11-25T17:01:38Z)
• MA-FSAR: Multimodal Adaptation of CLIP for Few-Shot Action Recognition [41.78245303513613]
  We introduce MA-FSAR, a framework that employs the Fine-Tuning (PEFT) technique to enhance the CLIP visual encoder in terms of action-related temporal and semantic representations. In addition to these token-level designs, we propose a prototype-level text-guided construction module to further enrich the temporal and semantic characteristics of video prototypes.
  arXiv  Detail & Related papers  (2023-08-03T04:17:25Z)
• GPT4Image: Large Pre-trained Models Help Vision Models Learn Better on Perception Task [47.1857510710807]
  We present a new learning framework, dubbed GPT4Image, where the knowledge of the large pre-trained models are extracted to help CNNs and ViTs learn better representations.<n>We conduct extensive experiments to verify the effectiveness of the proposed algorithm on various visual perception tasks.
  arXiv  Detail & Related papers  (2023-06-01T14:02:45Z)
• Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models [71.11425812806431]
  Latent Diffusion Models (LDMs) enable high-quality image synthesis while avoiding excessive compute demands. Here, we apply the LDM paradigm to high-resolution generation, a particularly resource-intensive task. We focus on two relevant real-world applications: Simulation of in-the-wild driving data and creative content creation with text-to-video modeling.
  arXiv  Detail & Related papers  (2023-04-18T08:30:32Z)

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.