MoCA-Video: Motion-Aware Concept Alignment for Consistent Video Editing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918245114576896 |
|---|---|
| author | Zhang, Tong Alcazar, Juan C Leon Escorcia, Victor Ghanem, Bernard |
| author_facet | Zhang, Tong Alcazar, Juan C Leon Escorcia, Victor Ghanem, Bernard |
| contents | We present MoCA-Video, a training-free framework for semantic mixing in videos. Operating in the latent space of a frozen video diffusion model, MoCA-Video utilizes class-agnostic segmentation with diagonal denoising scheduler to localize and track the target object across frames. To ensure temporal stability under semantic shifts, we introduce momentum-based correction to approximate novel hybrid distributions beyond trained data distribution, alongside a light gamma residual module that smooths out visual artifacts. We evaluate model's performance using SSIM, LPIPS, and a proposed metric, \metricnameabbr, which quantifies semantic alignment between reference and output. Extensive evaluation demonstrates that our model consistently outperforms both training-free and trained baselines, achieving superior semantic mixing and temporal coherence without retraining. Results establish that structured manipulation of diffusion noise trajectories enables controllable and high-quality video editing under semantic shifts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_01004 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MoCA-Video: Motion-Aware Concept Alignment for Consistent Video Editing Zhang, Tong Alcazar, Juan C Leon Escorcia, Victor Ghanem, Bernard Computer Vision and Pattern Recognition Artificial Intelligence We present MoCA-Video, a training-free framework for semantic mixing in videos. Operating in the latent space of a frozen video diffusion model, MoCA-Video utilizes class-agnostic segmentation with diagonal denoising scheduler to localize and track the target object across frames. To ensure temporal stability under semantic shifts, we introduce momentum-based correction to approximate novel hybrid distributions beyond trained data distribution, alongside a light gamma residual module that smooths out visual artifacts. We evaluate model's performance using SSIM, LPIPS, and a proposed metric, \metricnameabbr, which quantifies semantic alignment between reference and output. Extensive evaluation demonstrates that our model consistently outperforms both training-free and trained baselines, achieving superior semantic mixing and temporal coherence without retraining. Results establish that structured manipulation of diffusion noise trajectories enables controllable and high-quality video editing under semantic shifts. |
| title | MoCA-Video: Motion-Aware Concept Alignment for Consistent Video Editing |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2506.01004 |