AI Paper Atlas

Research map / Video / Video generation

Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models

Songwei Ge, Seungjun Nah, Guilin Liu, Tyler Poon, Andrew Tao, Bryan Catanzaro, David Jacobs, Jia-Bin Huang, Ming-Yu Liu, Yogesh Balaji

ICCV 2023 · Video › Video generation · approach: Efficient video diffusion models · 176 citations (OpenAlex)

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 changedWhereMeasuredEffect
finetune from a pretrained image diffusion model instead of training the video model from scratch (ablation)video diffusion on UCF-101 · UCF-101 small-scaleFVD: 903.37 -> 566.67Improved
replace i.i.d. frame noise with mixed correlated noise during finetuning (ablation)video diffusion finetuning · UCF-101 small-scaleFVD: 566.67 -> 337.40Improved
replace i.i.d. frame noise with progressive noise during finetuning (ablation)video diffusion finetuning · UCF-101 small-scaleIS: 41.25 -> 53.52Improved
use correlated noise to preserve image model quality (ablation)video diffusion finetuning · UCF-101 small-scaleframe FID: 56.43 -> 31.57Improved
use progressive instead of mixed noise (ablation)PYoCo · UCF-101 small-scaleFVD: 337.40 -> 339.67No clear effect
set correlation ratio alpha to 0 (i.i.d. noise) (ablation)PYoCo · UCF-101 small (69M)visual quality and diversityHurt
set correlation ratio too large (alpha=10) (ablation)PYoCo · UCF-101 small (69M)motion qualityHurt
use ensemble expert denoisers on top of base config (ablation)PYoCo · MSR-VTT zero-shotFID: 25.39 -> 22.14Improved
increase model size (69M to 253M) (ablation)PYoCo · UCF-101 small-scaleFVD: 332 -> 310 (112M to 253M)Improved
use PYoCo against Make-A-Video for zero-shot generationPYoCo · UCF-101 zero-shotInception Score: 33.45 -> 47.76Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most 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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy