P-Video-Replace

Image-to-Video • Pruna AI

Pruna's P-Video-Replace takes a source video and one or more identity reference images, then places the referenced person or people into the video while preserving the source motion and audio.

Model Info
More information	link ↗
Pricing	View pricing in the Cloudflare dashboard ↗

const response = await env.AI.run(
  'pruna/p-video-replace',
  {
    video: 'https://test-videos.co.uk/vids/bigbuckbunny/mp4/h264/360/Big_Buck_Bunny_360_10s_1MB.mp4',
    images: ['https://huggingface.co/spaces/yisol/IDM-VTON/resolve/main/example/human/00121_00.jpg'],
    resolution: '720p',
    target_fps: 'original',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "pruna/p-video-replace",
  "input": {
    "video": "https://test-videos.co.uk/vids/bigbuckbunny/mp4/h264/360/Big_Buck_Bunny_360_10s_1MB.mp4",
    "images": [
      "https://huggingface.co/spaces/yisol/IDM-VTON/resolve/main/example/human/00121_00.jpg"
    ],
    "resolution": "720p",
    "target_fps": "original"
  }
}'

Output
Raw response

{
  "state": "Completed",
  "result": {
    "video": "https://examples.aig.cloudflare.com/pruna/p-video-replace/character-swap.mp4"
  },
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Parameters

Input
Output

video

stringrequiredSource RGB video (.mp4) used as the motion and audio source. HTTP(S) URL or data URI.

▶images[]

arrayrequiredminItems: 1maxItems: 3Identity reference image(s), 1 to 3, to place into the video. Each entry is an HTTP(S) URL or a data URI.

turbo

booleanrequireddefault: falseTurbo mode: faster generation for slightly lower quality.

resolution

stringrequireddefault: 720penum: 720p, 1080pTarget resolution.

save_audio

booleanrequireddefault: trueSave the video with audio.

ignore_audio

booleanrequireddefault: falseIgnore source audio during generation.

target_fps

stringrequireddefault: originalenum: 24, 48, originalTarget FPS for the working video.

instruction_prompt

stringrequireddefault: Further instruction on how to place people from the reference images into the scene.

seed

integerminimum: -9007199254740991maximum: 9007199254740991Random seed for reproducible generation.

disable_safety_checker

booleanrequireddefault: falseDisable safety checker for generated videos.

video

stringformat: uriPresigned URL for the replacement video.

API Schemas (Raw)

Input

Output