CustomVideoX: 3D Reference Attention Driven Dynamic Adaptation for Zero-Shot Customized Video Diffusion Transformers

• URL: http://arxiv.org/abs/2502.06527v2
• Date: Thu, 20 Feb 2025 02:55:52 GMT
• Title: CustomVideoX: 3D Reference Attention Driven Dynamic Adaptation for Zero-Shot Customized Video Diffusion Transformers
• Authors: D. She, Mushui Liu, Jingxuan Pang, Jin Wang, Zhen Yang, Wanggui He, Guanghao Zhang, Yi Wang, Qihan Huang, Haobin Tang, Yunlong Yu, Siming Fu,
• Abstract summary: CustomVideoX capitalizes on pre-trained video networks by exclusively training the LoRA parameters to extract reference features.<n>We propose 3D Reference Attention, which enables direct and simultaneous engagement of reference image features.<n> Experimental results show that CustomVideoX significantly outperforms existing methods in terms of video consistency and quality.
• Score: 15.558659099600822
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Customized generation has achieved significant progress in image synthesis, yet personalized video generation remains challenging due to temporal inconsistencies and quality degradation. In this paper, we introduce CustomVideoX, an innovative framework leveraging the video diffusion transformer for personalized video generation from a reference image. CustomVideoX capitalizes on pre-trained video networks by exclusively training the LoRA parameters to extract reference features, ensuring both efficiency and adaptability. To facilitate seamless interaction between the reference image and video content, we propose 3D Reference Attention, which enables direct and simultaneous engagement of reference image features with all video frames across spatial and temporal dimensions. To mitigate the excessive influence of reference image features and textual guidance on generated video content during inference, we implement the Time-Aware Reference Attention Bias (TAB) strategy, dynamically modulating reference bias over different time steps. Additionally, we introduce the Entity Region-Aware Enhancement (ERAE) module, aligning highly activated regions of key entity tokens with reference feature injection by adjusting attention bias. To thoroughly evaluate personalized video generation, we establish a new benchmark, VideoBench, comprising over 50 objects and 100 prompts for extensive assessment. Experimental results show that CustomVideoX significantly outperforms existing methods in terms of video consistency and quality.

Related papers

• Subject-driven Video Generation via Disentangled Identity and Motion [52.54835936914813]
  We propose to train a subject-driven customized video generation model through decoupling the subject-specific learning from temporal dynamics in zero-shot without additional tuning. Our method achieves strong subject consistency and scalability, outperforming existing video customization models in zero-shot settings.
  arXiv  Detail & Related papers  (2025-04-23T06:48:31Z)
• Get In Video: Add Anything You Want to the Video [48.06070610416688]
  Video editing increasingly demands the ability to incorporate specific real-world instances into existing footage. Current approaches fail to capture the unique visual characteristics of particular subjects and ensure natural instance/scene interactions. We introduce "Get-In-Video Editing", where users provide reference images to precisely specify visual elements they wish to incorporate into videos.
  arXiv  Detail & Related papers  (2025-03-08T16:27:53Z)
• Multi-subject Open-set Personalization in Video Generation [110.02124633005516]
  We present Video Alchemist $-$ a video model with built-in multi-subject, open-set personalization capabilities.<n>Our model is built on a new Diffusion Transformer module that fuses each conditional reference image and its corresponding subject-level text prompt.<n>Our method significantly outperforms existing personalization methods in both quantitative and qualitative evaluations.
  arXiv  Detail & Related papers  (2025-01-10T18:59:54Z)
• MotionAura: Generating High-Quality and Motion Consistent Videos using Discrete Diffusion [3.7270979204213446]
  We present four key contributions to address the challenges of video processing. First, we introduce the 3D Inverted Vector-Quantization Variencoenco Autocoder. Second, we present MotionAura, a text-to-video generation framework. Third, we propose a spectral transformer-based denoising network. Fourth, we introduce a downstream task of Sketch Guided Videopainting.
  arXiv  Detail & Related papers  (2024-10-10T07:07:56Z)
• TC-Bench: Benchmarking Temporal Compositionality in Text-to-Video and Image-to-Video Generation [97.96178992465511]
  We argue that generated videos should incorporate the emergence of new concepts and their relation transitions like in real-world videos as time progresses. To assess the Temporal Compositionality of video generation models, we propose TC-Bench, a benchmark of meticulously crafted text prompts, corresponding ground truth videos, and robust evaluation metrics.
  arXiv  Detail & Related papers  (2024-06-12T21:41:32Z)
• Inflation with Diffusion: Efficient Temporal Adaptation for Text-to-Video Super-Resolution [19.748048455806305]
  We propose an efficient diffusion-based text-to-video super-resolution (SR) tuning approach. We investigate different tuning approaches based on our inflated architecture and report trade-offs between computational costs and super-resolution quality.
  arXiv  Detail & Related papers  (2024-01-18T22:25:16Z)
• MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing [90.06041718086317]
  We propose a unified Multi-alignment Diffusion, dubbed as MagDiff, for both tasks of high-fidelity video generation and editing. The proposed MagDiff introduces three types of alignments, including subject-driven alignment, adaptive prompts alignment, and high-fidelity alignment.
  arXiv  Detail & Related papers  (2023-11-29T03:36:07Z)
• VideoGen: A Reference-Guided Latent Diffusion Approach for High Definition Text-to-Video Generation [73.54366331493007]
  VideoGen is a text-to-video generation approach, which can generate a high-definition video with high frame fidelity and strong temporal consistency. We leverage an off-the-shelf text-to-image generation model, e.g., Stable Diffusion, to generate an image with high content quality from the text prompt.
  arXiv  Detail & Related papers  (2023-09-01T11:14:43Z)
• Control-A-Video: Controllable Text-to-Video Diffusion Models with Motion Prior and Reward Feedback Learning [50.60891619269651]
  Control-A-Video is a controllable T2V diffusion model that can generate videos conditioned on text prompts and reference control maps like edge and depth maps. We propose novel strategies to incorporate content prior and motion prior into the diffusion-based generation process. Our framework generates higher-quality, more consistent videos compared to existing state-of-the-art methods in controllable text-to-video generation.
  arXiv  Detail & Related papers  (2023-05-23T09:03:19Z)
• A Video Is Worth Three Views: Trigeminal Transformers for Video-based Person Re-identification [77.08204941207985]
  Video-based person re-identification (Re-ID) aims to retrieve video sequences of the same person under non-overlapping cameras. We propose a novel framework named Trigeminal Transformers (TMT) for video-based person Re-ID.
  arXiv  Detail & Related papers  (2021-04-05T02:50:16Z)
• Vid-ODE: Continuous-Time Video Generation with Neural Ordinary Differential Equation [42.85126020237214]
  We propose continuous-time video generation by combining neural ODE (Vid-ODE) with pixel-level video processing techniques. Vid-ODE is the first work successfully performing continuous-time video generation using real-world videos.
  arXiv  Detail & Related papers  (2020-10-16T06:50:47Z)

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.