AI Paper Atlas

Research map / Video / Video generation

LEO: Generative Latent Image Animator for Human Video Synthesis

Yaohui Wang, Xin Ma, Xinyuan Chen, Cunjian Chen, Antitza Dantcheva, Bo Dai, Yu Qiao

IJCV 2024 · Video › Video generation · approach: Trajectory-guided motion control · 42 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 Linear Motion Condition (LMC) to LMDM (ablation)LEO · TaichiHDFVD: 118.6 -> 100.4Improved
add Linear Motion Condition (LMC) to LMDM (ablation)LEO · FaceForensicsFVD: 60.03 -> 52.32Improved
add Transition Diffusion Model for long TaichiHD videos (ablation)LEO · TaichiHD long videorepeated motion occurrence (%): 0.45 -> 0.02Improved
condition on a real starting image instead of cDDPM-generated oneLEO · TaichiHD128FVD16: 100.4 -> 57.6Improved
condition on a real starting imageLEO · FaceForensicsFVD16: 52.3 -> 35.9Improved
compare LEO to DIGAN at 256x256 resolutionLEO (uncond) · TaichiHD256FVD16: 156.7 -> 122.7Improved
generate 128-frame videos with LEO instead of TATSLEO · TaichiHD128 longFVD128: 1194.58 -> 155.54Improved
generate 128-frame videos with LEO instead of StyleGAN-VLEO · FaceForensics longFVD128: 89.34 -> 96.28Hurt
unconditional LEO versus StyleGAN-V on FaceForensics short videosLEO (uncond) · FaceForensicsFVD16: 47.4 -> 52.3Hurt
apply LEO to general videos versus large diffusion modelsLEO · UCF101FVD16: 367.23 -> 1356.2Hurt
apply LEO instead of StyleGAN-V on UCF101LEO · UCF101FVD16: 1431.0 -> 1356.2Improved

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