AI Paper Atlas

Research map / Video / Video generation

MVI2V: Human Centric Image to Video Generation with Multiview Consistent Appearance

Pengfei Liu, Mingyi Xu, Wentao Jiang, Tiezheng Ge, Ming Zeng

ECCV 2026 · Video › Video generation · approach: Trajectory-guided motion control · 0 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 MVI2V architecture (dedicated reference stream) (ablation)Wan2.1 · Multi-view benchmarkGPT garment: 0.409 -> 0.470Improved
add reference frame selection (ablation)MVI2V-Wan2.1 · Multi-view benchmarkGPT garment: 0.470 -> 0.562Improved
add reference frame selection (ablation)MVI2V-Wan2.1 · Multi-view benchmarkGPT person: 0.784 -> 0.853Improved
add inpainting subtask (ablation)MVI2V-Wan2.1 · Multi-view benchmarkGPT garment: 0.562 -> 0.585Improved
add inpainting subtask (ablation)MVI2V-Wan2.1 · Multi-view benchmarkDINO person: 0.743 -> 0.728Hurt
add inpainting subtask (ablation)MVI2V-Wan2.1 · Multi-view benchmarkDINO garment: 0.638 -> 0.656Improved
use dedicated reference stream instead of Token-ConcatWan2.1 · Multi-view benchmarkGPT garment: 0.512 -> 0.585Improved
use dedicated reference stream instead of Concat-IDWan2.1 · Multi-view benchmarkGPT person: 0.841 -> 0.862Improved
add Token-Concat multi-view referencesWan2.1 · Multi-view benchmarkSubject Consistency: 0.916 -> 0.902Hurt
add time projection layer for reference tokens (Concat-ID vs Token-Concat)Token-Concat · Multi-view benchmarkSubject Consistency: 0.902 -> 0.918Improved
add MVI2V triple-stream extensionIn-House · Multi-view benchmarkGPT person: 0.692 -> 0.868Improved
add MVI2V triple-stream extensionIn-House · Multi-view benchmarkDINO garment: 0.601 -> 0.633Improved

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