AI Paper Atlas

Research map / Video / Video generation

Mobile Video Diffusion

Haitam Ben Yahia, Denis Korzhenkov, Ioannis Lelekas, Amir Ghodrati, Amirhossein Habibian

ICCV 2025 · Video › Video generation · approach: Efficient video diffusion models · 17 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
low-resolution finetuning of SVD at 512x256 (ablation)SVD · UCF-101 7 FPSFVD: 476 -> 196Improved
add one-step adversarial finetuning (ablation)SVD at 512x256 · UCF-101 7 FPSFVD: 196 -> 168Improved
remove no-op query/key similarity in cross-attention (ablation)SVD UNet · Xiaomi-14 Prophone latency: OOM -> 3630 msImproved
use temporal multiscaling (ablation)MobileVD UNet · UCF-101 7 FPSphone latency: 3630 -> 2590 ms; FVD 168 -> 156Improved
use spatial instead of temporal multiscaling (ablation)MobileVD UNet · UCF-101FVD: 145 vs 138 FVDHurt
combine spatial and temporal multiscaling (ablation)MobileVD UNet · UCF-101FVD: 133 -> 163Hurt
add learned temporal block pruning (70%) (ablation)MobileVD · UCF-101 7 FPSphone latency: 2590 -> 2100 ms; FVD 156 -> 150Improved
prune temporal blocks to 80% instead of 70% (ablation)MobileVD · UCF-101FVD: 127 -> 245Hurt
use proposed learned pruning instead of L1 regularization (ablation)temporal block pruning at 70% · UCF-101FVD: 207 -> 127Improved
add channel funnels (fun-factor 0.5) (ablation)MobileVD · UCF-101phone latency: 2100 -> 1780 ms; FVD 150 -> 171Improved
use coupled singular initialization instead of He init for funnels (ablation)channel funnels · UCF-101FVD: 332 -> 132Improved
reduce fun-factor to 0.25 (ablation)channel funnels · UCF-101FVD: 133 -> 155Hurt

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