Research map / Image generation / Fast diffusion sampling
DreamMover: Leveraging the Prior of Diffusion Models for Image Interpolation with Large Motion
Liao Shen, Tianqi Liu, Huiqiang Sun, Xinyi Ye, Baopu Li, Jianming Zhang, Zhiguo Cao
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 two-level fusion (high-level and low-level spaces) instead of fusing in z_T space (ablation) | DreamMover · InterpBench | FID: 138.9329 -> 43.1798 | Improved |
| add the diffusion-based flow estimator instead of off-the-shelf RAFT (ablation) | DreamMover · InterpBench | FID: 53.8510 -> 43.1798 | Improved |
| add self-attention replacement (ablation) | DreamMover · InterpBench | FID: 69.4221 -> 43.1798 | Improved |
| add LoRA fine-tuned on the image pair (ablation) | DreamMover · InterpBench | FID: 76.7142 -> 43.1798 | Improved |
| use DreamMover instead of video interpolation methods Film and LDMVFI | large-motion image interpolation · InterpBench | Warping Error: 0.1244 (Film) vs 0.2069 | Hurt |
| use DreamMover instead of best baseline LDMVFI | large-motion image interpolation · InterpBench | FID: 48.3469 -> 43.1798 | Improved |
| use DreamMover instead of Film in user study | large-motion image interpolation · InterpBench user study | human preference: 24.77% -> 75.23% | Improved |
Closest papers by meaning
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.88)
- Dual Diffusion for Unified Image Generation and Understanding (CVPR 2025 · similarity 0.88)
- Spin: Diffusion-based Semantic Image Painting Through Independent Information Injection (AAAI 2025 · similarity 0.87)
- AID: Attention Interpolation of Text-to-Image Diffusion (NeurIPS 2024 · similarity 0.87)
- Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models (ICLR 2024 · similarity 0.87)
- Emergent Temporal Correspondences from Video Diffusion Transformers (NeurIPS 2025 · similarity 0.87)
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.
