Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Subject-driven video generation research benchmarks · SOTA2 Research

Subject-driven video generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

98.93Motion Smoothness

8

May 6, 2026

Subject-driven Video Generation

10Training Steps

7

May 6, 2026

0.896CLIP-I

5

Mar 20, 2026

Subject-to-Video (S2V) (test)

0.538MS

5

Mar 17, 2026

Human-Object Interaction (HOI) scenes

71.7DINO (v -> r)

5

Feb 26, 2026

Object-Centric (OC) scenes

0.816DINO Sim (V->R)

5

Feb 26, 2026

S2V 4 reference images (test)

70GSB Ratio

4

Mar 17, 2026

S2V 3 reference images (test)

0.429GSB Ratio

4

Mar 17, 2026

S2V 2 reference images (evaluation set)

0.62GSB Ratio

4

Mar 17, 2026

S2V 1 reference image (val)

42.3GSB Ratio

4

Mar 17, 2026

Popular datasets

S2V VBench Object-Centric scenes Human-Object Interaction (HOI)Subject-to-Video (S2V)

Follow for updates

Join SOTA2 Contact