Research map / Video / Video generation
Motion Prior Distillation in Time Reversal Sampling for Generative Inbetweening
Wooseok Jeon, Seunghyun Shin, Dongmin Shin, Hae-Gon Jeon
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 Motion Prior Distillation to ViBiD sampling | ViBiD (SVD-XT) · DAVIS | FVD: 559.49 -> 527.05 | Improved |
| add Motion Prior Distillation to TRF sampling | TRF (SVD-XT) · DAVIS | FID: 56.894 -> 34.910 | Improved |
| add Motion Prior Distillation to TRF sampling | TRF (SVD-XT) · Pexels | FVD: 796.48 -> 460.99 | Improved |
| add MPD to ViBiD sampling | ViBiD (SVD-XT) · Pexels | VBench: 0.8130 -> 0.8235 | Mixed |
| compare with FILM on VBench++ | Ours + TRF · DAVIS | VBench++: 0.9740 vs 0.9330 | Hurt |
| apply MPD at all denoising steps instead of only early steps (ablation) | Ours + ViBiD · DAVIS | FVD: 527.05 -> 1086.6 | Hurt |
| increase re-noising steps k from 1 to 3 (ablation) | Ours + ViBiD · DAVIS | FVD: 568.06 -> 527.05 | Improved |
| increase re-noising steps k from 1 to 3 (ablation) | Ours + TRF · DAVIS | FVD: 588.27 -> 662.75 | Hurt |
| use interpolation scale 1.0 instead of 0.5 (ablation) | Ours + ViBiD · DAVIS | FVD: 539.99 -> 527.05 | Improved |
| use interpolation scale 1.0 instead of 0.5 (ablation) | Ours + TRF · DAVIS | FVD: 612.17 -> 654.72 | Hurt |
| reduce distillation ratio gamma from 0.3 to 0.1 (ablation) | Ours + TRF · DAVIS | FVD: 612.17 -> 573.13 | Mixed |
Closest papers by meaning
- Generative Inbetweening: Adapting Image-to-Video Models for Keyframe Interpolation (ICLR 2025 · similarity 0.91)
- Motion Consistency Model: Accelerating Video Diffusion with Disentangled Motion-Appearance Distillation (NeurIPS 2024 · similarity 0.90)
- Identifying and Solving Conditional Image Leakage in Image-to-Video Diffusion Model (NeurIPS 2024 · similarity 0.89)
- TRIP: Temporal Residual Learning with Image Noise Prior for Image-to-Video Diffusion Models (CVPR 2024 · similarity 0.89)
- Generative Pre-trained Autoregressive Diffusion Transformer (NeurIPS 2025 · similarity 0.88)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.88)
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.
