AI Paper Atlas

Research map / Video / Video generation

SA-V2V: Training-Free Subject-Aware Video-to-Video Personalization

Soobin Park, Seohyeon Yoo, Jiwon Kim, SeonHwa Kim, Kyong Hwan Jin, Eunju Cha

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 stochastic temporal injection (ablation)SA-V2V · Video personalization (ablation set)FVD: 103.82 -> 82.84Improved
add stochastic temporal injection (ablation)SA-V2V · Video personalization (ablation set)Masked CLIP background similarity: 0.8146 -> 0.8783Improved
add background target-aware feature guidance (B-TFG) (ablation)SA-V2V · Video personalization (ablation set)Masked CLIP background similarity: 0.8313 -> 0.8783Improved
add background target-aware feature guidance (B-TFG) (ablation)SA-V2V · Video personalization (ablation set)FVD: 97.27 -> 82.84Improved
add subject target-aware feature guidance (S-TFG) (ablation)SA-V2V · Video personalization (ablation set)Masked CLIP subject similarity: 0.8101 -> 0.8312Improved
add subject target-aware feature guidance (S-TFG) (ablation)SA-V2V · Video personalization (ablation set)Masked CLIP background similarity: 0.9041 -> 0.8783Hurt
add subject target-aware feature guidance (S-TFG) (ablation)SA-V2V · Video personalization (ablation set)FVD: 72.30 -> 82.84Hurt
add spatial masking to TFG (ablation)SA-V2V TFG · qualitative (Fig. 7)feature separationImproved
apply TFG alternately timestep-wise instead of block-wise (ablation)SA-V2V TFG · qualitative (Fig. 7)appearance guidanceImproved
use linear interpolation instead of direct spatial feature injection (ablation)SA-V2V TFG · qualitative (Fig. 7)background preservationImproved
use training-free design instead of training-based baselinesSA-V2V · Computational cost (CogVideoX-5B)Peak GPU memory (GB): 21.72 -> 19.11Improved
SA-V2V vs MotionClone on subject fidelitySA-V2V · Video personalization (main comparison)Masked CLIP subject similarity: 0.8243 -> 0.8140Hurt

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