Dreamactor V2

Dreamactor V2 animates a character image from a reference performance video, transferring facial expressions, head motion, and body gestures to photographic or illustrated subjects.

Video Motion Control

Based on verified input duration.

Offering	Price (USD)
Usage rate	$0.05 / input second

Based on the video input; billing steps apply in this order:
Round input duration up to a whole second.
Cap billed duration at 30 seconds.
Round the total charge up to the next $0.01.

Dreamactor V2

Dreamactor V2 produces character animation from a still image and a driving video. The image establishes the subject's appearance; the video provides the performance, including facial expressions, head movement, and body gestures. The generated motion follows the timing of the recorded actions.

Photographic portraits, illustrations, and anime-style characters can serve as the visual source. This makes a recorded human performance useful for exploring how a drawn or photographed character moves and expresses emotion.

What it's good for

Character artists can use it to test an expressive performance on a static design. A reference clip showing a reaction or a sequence of gestures gives the animation a concrete performance to follow.

It also suits animated mascots and avatar-led social clips. For entertainment projects, teams can use a performer's movements to develop character scenes, connecting existing artwork to a recorded performance during creative development.

Model ID

bytedance/dreamactor-v2

Output Type

Video

Pricing

$0.05 / input second

Provider

ByteDance

Hosting

Passthru

Dreamactor V2

What it's good for

Character artists can use it to test an expressive performance on a static design. A reference clip showing a reaction or a sequence of gestures gives the animation a concrete performance to follow.