Research map / Video / Video generation
VIDM: Video Implicit Diffusion Models
Kangfu Mei, Vishal Patel
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 |
|---|---|---|---|
| use sampling space truncation (ablation) | VIDM content generator · 256-CLEVRER 16 frames | FID: 18.4 -> 21.1 | Improved |
| use robustness penalty (ablation) | VIDM content generator · 256-CLEVRER 16 frames | FVD: 95.5 -> 87.4 | Improved |
| add PosGN (ablation) | VIDM motion generator · 256-CLEVRER 16 frames | FVD: 532.1 -> 87.4 | Improved |
| add PosGN (ablation) | VIDM motion generator · 256-CLEVRER 64 frames | FVD: 581.3 -> 286.6 | Improved |
| add PosGN (ablation) | VIDM motion generator · 256-CLEVRER 128 frames | FVD: 604.5 -> 426.5 | Improved |
| add implicit motion conditions (ablation) | VIDM motion generator · 256-CLEVRER 16 frames | FVD: 584.8 -> 87.4 | Improved |
| add implicit motion conditions (ablation) | VIDM motion generator · 256-CLEVRER 64 frames | FVD: 552.1 -> 286.6 | Improved |
| add implicit motion conditions (ablation) | VIDM motion generator · 256-CLEVRER 128 frames | FVD: 614.1 -> 426.5 | Improved |
| add positional encoding and implicit latent conditioning to diffusion video generation (ablation) | VIDM motion generator · long video generation | motion continuity | Improved |
| apply the robustness penalty (ablation) | VIDM content and motion generators · video generation | modeling ability | Improved |
Later papers that compared with this method
2 later accepted papers report VIDM as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- FIFO-Diffusion: Generating Infinite Videos from Text without Training (NeurIPS 2024 · better on 1 of 1 shared result)
- GLOBER: Coherent Non-autoregressive Video Generation via GLOBal Guided Video DecodER (NeurIPS 2023 · better on 5 of 5 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- StyleGAN-V — StyleGAN-V: A Continuous Video Generator With the Price, Image Quality and Perks of StyleGAN2 (CVPR 2022) · 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
- SF-V: Single Forward Video Generation Model (NeurIPS 2024 · similarity 0.91)
- Video Interpolation with Diffusion Models (CVPR 2024 · similarity 0.90)
- OSV: One Step is Enough for High-Quality Image to Video Generation (CVPR 2025 · similarity 0.90)
- Diffusion Adversarial Post-Training for One-Step Video Generation (ICML 2025 · similarity 0.89)
- Efficient Video Diffusion Models via Content-Frame Motion-Latent Decomposition (ICLR 2024 · similarity 0.89)
- Grid Diffusion Models for Text-to-Video Generation (CVPR 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.
