Video examples

Dime Dime Baby video examples

Compare the input, movement and final frames, with the source and conditions recorded for each published example.

Original AI demonstrations

Fictional-adult first-frame demonstrations with the original input, complete output and observed limits.

AI-generated first frame of a fictional adult in an orange overshirt holding a bread roll in a warm party room
AI-generated first frame
Original AI demonstrations · 8.04s · Silent MP4

Original AI demonstrations

Historical Mini first-frame experiment

A historical Seedance 2 Mini experiment using an original AI-generated first frame and independent motion instructions. A fictional adult in an orange overshirt approaches the camera, brings a bread roll to his mouth, looks toward the lens, then turns and walks back.

Conditions

  • The input image, central adult and background guests were newly generated from text. The depicted people are fictional adults.
  • Seedance 2 Mini through Kie.ai received that image as the first frame, together with independently written motion instructions. The request supplied no reference video or reference audio.
  • The request used eight seconds, the 480p preset, 4:3 framing and audio generation off. The completed MP4 is 752 × 560 pixels at 24 frames per second and lasts about 8.04 seconds.

What to look for

  • Around 0.5–1.5 seconds, the adult steps toward the camera while holding the same roll in his right hand.
  • Around 1.5–2.5 seconds, the roll reaches his mouth. He then lowers his hand, appears to chew and looks toward the camera.
  • Around 5.5–6.5 seconds, he turns. During the final part of the clip, he walks back toward the guests with his back facing the camera.
  • The orange overshirt, light T-shirt and room remain recognizable across the sequence. Pause during the mouth contact and turn to inspect the generated details.

Limits of this example

  • This historical first-frame experiment uses a different workflow from the current photo-and-motion-reference template. It does not demonstrate identity transfer from your portrait in the current generator.
  • The bite happens earlier than the requested timing. Facial detail, fingers, the bread surface and background details vary slightly between frames.
  • The MP4 is silent. Music added for publication requires permission for the chosen platform and intended use.
  • This is one observed output. The input and generated media are disclosed as AI-generated, and neither is a customer portrait or a customer submission.

ByteDance Seedance 2 Mini via Kie.ai · 480p preset · actual 752 × 560 ·Method first-frame-image-to-video

Seedance 2 Mini model information (external source)Kie.ai terms of use (external source)OpenAI terms for generated content (external source)Open the generator