AI Paper Atlas

Research map / Video / Video generation

OmniShow: Unifying Multimodal Conditions for Human-Object Interaction Video Generation

Donghao Zhou, Guisheng Liu, Hao Yang, Jiatong Li, Jingyu Lin, Xiaohu Huang, Yichen Liu, Xin Gao, Cunjian Chen, Shilei Wen, Chi-Wing Fu, Pheng-Ann Heng

ICML 2026 · Video › Video generation · approach: Diffusion-based human image animation · 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 unified channel-wise conditioning instead of token concatenation (ablation)OmniShow · Ablation (a) R2VFaceSim: 0.601 -> 0.707Improved
add reference reconstruction loss (ablation)OmniShow · Ablation (a) R2VFaceSim: 0.678 -> 0.707Improved
add attention-map constraints in gated local-context attention (ablation)OmniShow · Ablation (b) RA2VSync-C: 2.201 -> 9.023Improved
add audio context (local window) (ablation)OmniShow · Ablation (b) RA2VSync-D: 7.878 -> 7.419Improved
add adaptive gating (ablation)OmniShow · Ablation (b) RA2VSync-D: 7.819 -> 7.419Improved
add attention-map constraints (ablation)OmniShow · Ablation (b) RA2VAES: 0.545 -> 0.540Hurt
use decoupled-then-joint training instead of single-stage RA2V (ablation)OmniShow · Ablation (c) RA2VNexusScore: 0.345 -> 0.364Improved
use decoupled-then-joint training instead of R2V->RA2V multi-stage (ablation)OmniShow · Ablation (c) RA2VSync-D: 13.23 -> 8.14Improved
pretrain on A2V then RA2V instead of ours (ablation)OmniShow · Ablation (c) RA2VSync-D: 7.38 vs 8.14No clear effect
use native RoPE instead of spatiotemporal shift for pseudo-frames (ablation)OmniShow · R2VFaceSim: 0.279 -> 0.707Improved
use native RoPE instead of temporal shift (ablation)OmniShow · R2VFaceSim: 0.675 -> 0.707Improved
use context window 5 instead of 11 (ablation)OmniShow · RA2VSync-C: 7.020 -> 9.023Improved

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