A drum head is struck, producing a bright tone. Then water is sprinkled on the surface and it is struck again, sounding lower and deader.
Seedance 2.0ByteDance · proprietary
Kling 3.0 OmniKuaishou · proprietary
Veo 3.1Google · proprietary
LTX-2.3Lightricks · open
Ovi 1.1character.ai · open
JavisDiT++JavisDiT · open
MagiHumanGAIR-NLP · open
Per-prompt rubric
Basic standards
Visual presence
- a drum
- water
Event: a drum head is struck
Audio presence
- a drum
Sound: a drum hit
Key standards
Visual physical commonsense (V-PC)
- The drum head is visibly struck twice, and water is visibly sprinkled or poured onto the drum surface between the two strikes.
Audio physical commonsense (A-PC)
- The second hit with water on the head is noticeably lower and more damped than the first hit.
- Each strike sound has the characteristic drumhead impact timbre.
Cross-modal physical commonsense (AV-PC)
- Each hit sound aligns in time with a visible strike of the drum head.
- The lower and more damped hit corresponds to the drum after water is added.
- The brighter hit corresponds to the dry drum.