Research map / Video / Video generation
DualReal: Adaptive Joint Training for Lossless Identity-Motion Fusion in Video Customization
Wenchuan Wang, Mengqi Huang, Yijing Tu, Zhendong Mao
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 weight grouping (ablation) | DualReal · Ablation subset | DINO-I: 0.766 -> 0.771 | Improved |
| add weight grouping (ablation) | DualReal · Ablation subset | CLIP-I: 0.662 -> 0.674 | Improved |
| add weight grouping (ablation) | DualReal · Ablation subset | CLIP-T: 0.335 -> 0.333 | No clear effect |
| add first ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | DINO-I: 0.647 -> 0.771 | Improved |
| add first ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | Dynamic Degree: 3.51 -> 6.34 | Improved |
| add first ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | CLIP-T: 0.334 -> 0.333 | Hurt |
| add second ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | CLIP-I: 0.619 -> 0.674 | Improved |
| add second ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | CLIP-T: 0.346 -> 0.333 | Hurt |
| add second ablated component (Dual-aware Adaptation or StageBlender) (ablation) | DualReal · Ablation subset | Dynamic Degree: 5.70 -> 6.34 | Improved |
| use group cardinality n=7 instead of n=1 (ablation) | DualReal · Ablation subset, group cardinality | DINO-I: 0.766 -> 0.771 | Improved |
| use group cardinality n=7 instead of n=2 (ablation) | DualReal · Ablation subset, group cardinality | DINO-I: 0.660 -> 0.771 | Improved |
| use group cardinality n=7 instead of n=42 (ablation) | DualReal · Ablation subset, group cardinality | DINO-I: 0.706 -> 0.771 | Improved |
Later papers that compared with this method
3 later accepted papers report DualReal as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Composing Concepts from Images and Videos via Concept-prompt Binding (CVPR 2026 · better on 6 of 6 shared results)
- SMRABooth: Subject and Motion Representation Alignment for Customized Video Generation (CVPR 2026 · better on 5 of 5 shared results)
- SA-V2V: Training-Free Subject-Aware Video-to-Video Personalization (ECCV 2026 · better on 5 of 5 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- VideoMage: Multi-Subject and Motion Customization of Text-to-Video Diffusion Models (CVPR 2025 · similarity 0.88)
- VMC: Video Motion Customization using Temporal Attention Adaption for Text-to-Video Diffusion Models (CVPR 2024 · similarity 0.87)
- Identity-Preserving Text-to-Video Generation by Frequency Decomposition (CVPR 2025 · similarity 0.87)
- ARGUS: Stacked Multi-View Identity Mosaic Injection for Subject-Preserving Video Generation (NeurIPS 2026 · similarity 0.87)
- X-NeMo: Expressive Neural Motion Reenactment via Disentangled Latent Attention (ICLR 2025 · similarity 0.87)
- PolyVivid: Vivid Multi-Subject Video Generation with Cross-Modal Interaction and Enhancement (NeurIPS 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.
