AI Paper Atlas

Research map / Video / Video generation

MoGAN: Improving Motion Quality in Video Diffusion via Few-Step Motion Adversarial Post-Training

Haotian Xue, Qi Chen, Zhonghao Wang, Xun Huang, Eli Shechtman, Jinrong Xie, Yongxin Chen

ECCV 2026 · Video › Video generation · approach: Efficient video diffusion models · 7 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
add DMD regularization loss (ablation)MoGAN · Ablation (Table 2)motion score: 0.674 -> 0.983Improved
add DMD regularization loss (ablation)MoGAN · Ablation (Table 2)dynamics degree: 0.35 -> 0.98Improved
add DMD regularization loss (ablation)MoGAN · Ablation (Table 2)aesthetic: 0.50 -> 0.57Improved
add R1 and R2 discriminator regularization (ablation)MoGAN · Ablation (Table 2)smoothness (%): 95.1 -> 98.5Improved
add R1 and R2 discriminator regularization (ablation)MoGAN · Ablation (Table 2)motion score: 0.905 -> 0.983Improved
compute GAN loss in optical-flow space instead of video space (ablation)MoGAN · Ablation (Table 2)dynamics degree: 0.85 -> 0.98Improved
compute GAN loss in optical-flow space instead of video space (ablation)MoGAN · Ablation (Table 2)image quality: 0.67 -> 0.66No clear effect
distill 50-step Wan into 3-step with DMD onlyWan2.1-T2V-1.3B · VBenchdynamics degree: 0.83 -> 0.73Hurt
add motion GAN post-training to DMD 3-step modelDMD-only · VBenchmotion score: 0.859 -> 0.973Improved
add motion GAN post-training to DMD 3-step modelDMD-only · VideoJAM-Benchmotion score: 0.898 -> 0.983Improved
add motion GAN post-training to DMD 3-step modelDMD-only · VBenchdynamics degree: 0.73 -> 0.96Improved
add motion GAN post-training to DMD 3-step modelDMD-only · VBenchsmoothness (%): 98.8 -> 98.6Mixed

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