EM-Vid: Training-Free Entity-Centric Memory for Efficient and Consistent Multi-Shot Video Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914592190365696 |
|---|---|
| author | Vandersanden, Jente Gadelha, Matheus Huang, Chun-Hao P. Jeong, Hyeonho Gryaditskaya, Yulia |
| author_facet | Vandersanden, Jente Gadelha, Matheus Huang, Chun-Hao P. Jeong, Hyeonho Gryaditskaya, Yulia |
| contents | Multi-shot video generation requires maintaining a consistent appearance of recurring entities across shots while remaining faithful to shot-specific text prompts. Recent autoregressive methods reuse previously generated frames as memory. However, full-frame storage entangles persistent entity information with transient scene context, leading to irrelevant information leakage and high computational cost. We propose an entity-centric memory in the form of an entity-indexed bank of latent patches. We introduce sparse token conditioning compatible with pretrained models, restricting self-attention to entity-relevant tokens and reducing computational cost. To support this, we introduce a structured multi-shot script format. We additionally propose a budgeted memory update strategy to maintain a compact, evolving memory. Finally, we equip the entity representation with a noise-injection mechanism that enables fine-grained appearance control, preventing leakage of irrelevant information. Our method improves prompt adherence and efficiency while preserving subject consistency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_23610 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EM-Vid: Training-Free Entity-Centric Memory for Efficient and Consistent Multi-Shot Video Generation Vandersanden, Jente Gadelha, Matheus Huang, Chun-Hao P. Jeong, Hyeonho Gryaditskaya, Yulia Computer Vision and Pattern Recognition Artificial Intelligence Multi-shot video generation requires maintaining a consistent appearance of recurring entities across shots while remaining faithful to shot-specific text prompts. Recent autoregressive methods reuse previously generated frames as memory. However, full-frame storage entangles persistent entity information with transient scene context, leading to irrelevant information leakage and high computational cost. We propose an entity-centric memory in the form of an entity-indexed bank of latent patches. We introduce sparse token conditioning compatible with pretrained models, restricting self-attention to entity-relevant tokens and reducing computational cost. To support this, we introduce a structured multi-shot script format. We additionally propose a budgeted memory update strategy to maintain a compact, evolving memory. Finally, we equip the entity representation with a noise-injection mechanism that enables fine-grained appearance control, preventing leakage of irrelevant information. Our method improves prompt adherence and efficiency while preserving subject consistency. |
| title | EM-Vid: Training-Free Entity-Centric Memory for Efficient and Consistent Multi-Shot Video Generation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2605.23610 |