Research map / Video / Video generation
Bring Your Dreams to Life: Continual Text-to-Video Customization
Jiahua Dong, Xudong Wang, Wenqi Liang, Zongyan Han, Meng Cao, Duzhen Zhang, Hanbin Zhao, Zhi Han, Salman Khan, Fahad Shahbaz Khan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add task-aware concept aggregation to the baseline (ablation) | DreamVideo + LoRA-M baseline · CVC single-concept | CLIP-T: 25.5 -> 28.7 | Improved |
| add task-aware concept aggregation to the baseline (ablation) | baseline · CVC single-concept | F-CLIP-T (forgetting): 5.2 -> 3.4 | Improved |
| add layer-specific concept tokens on top of TCA (ablation) | CCVD · CVC single-concept | F-CLIP-T (forgetting): 3.4 -> 1.3 | Improved |
| add layer-specific concept tokens on top of TCA (ablation) | CCVD · CVC single-concept | CLIP-I: 62.2 -> 63.6 | Improved |
| add concept orthogonal loss (ablation) | CCVD · CVC single-concept | CLIP-I: 63.6 -> 64.3 | Improved |
| add concept orthogonal loss (ablation) | CCVD · CVC single-concept | F-CLIP-T (forgetting): 1.3 -> 1.1 | Improved |
| CCVD instead of best baseline L2DM | DreamVideo backbone · CVC single-concept | F-CLIP-T (forgetting): 3.1 -> 1.1 | Improved |
| CCVD instead of L2DM | DreamVideo backbone · CVC single-concept | TCons: 96.2 -> 96.4 | No clear effect |
| sequential plain finetuning without continual-learning strategy | DreamVideo · CVC single-concept | F-CLIP-T (forgetting): 9.5 vs 1.1 for CCVD | Hurt |
Closest papers by meaning
- How to Continually Adapt Text-to-Image Diffusion Models for Flexible Customization? (NeurIPS 2024 · similarity 0.88)
- CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities (AAAI 2025 · similarity 0.87)
- ConceptGuard: Continual Personalized Text-to-Image Generation with Forgetting and Confusion Mitigation (CVPR 2025 · similarity 0.87)
- StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text (CVPR 2025 · similarity 0.87)
- VideoTetris: Towards Compositional Text-to-Video Generation (NeurIPS 2024 · similarity 0.86)
- Mimir: Improving Video Diffusion Models for Precise Text Understanding (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
