Video Head Swap API Documentation

Video Head Swap API Documentation

Playground

Instant online AI head & face swap for videos with no watermark, delivering realistic, shareable results in seconds. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

WaveSpeedAI Video Head Swap is an advanced AI model for replacing the entire head (face + hair + outline) of a person in a video using a reference portrait. The model keeps the body, pose, and background intact while reconstructing a new, realistic head that matches the original lighting and perspective.

🎬 What this model does

• Replaces the full head region of the subject (face, hair, silhouette, accessories)
• Preserves body pose, clothing, background, and overall composition
• Adapts the new head to scene lighting, color tone, and camera angle
• Produces clean, watermark-free outputs ready for editing or publication

⚙️ Why it looks realistic

• Full-head geometry replacement Swaps the entire head contour instead of only facial features, avoiding mismatched hairlines or distorted skull shapes.

• Pose and expression preservation Follows the motion in the source video so head angle, gaze direction, and expression remain consistent with the original performance.

• Lighting and color matching Automatically adjusts skin tone, shadows, and highlights so the new head blends naturally into the scene.

• High-resolution blending Smooth transitions around hair, neck, and accessories, minimizing visible seams or flicker across frames.

💰 Pricing

> Billing duration: Input media duration is rounded up to whole seconds. The minimum billed duration is 3 seconds; shorter inputs are billed as 3 seconds. Existing per-5-second rate tables remain unchanged.

Pricing is based on video duration and output resolution, with a 3-second minimum and 120-second cap.

• minimum billed duration: 3 seconds
• Maximum billed duration: 120 seconds per run (longer clips are capped at 120 s)

🔧 Input Parameters

video (required)

The source video whose head you want to replace. This defines body motion, framing, and background.

face_image / head_image (required)

A clear portrait of the target identity. Frontal or three-quarter views with good lighting work best.

resolution

Output resolution for the processed video, for example:

• 480p – more affordable drafts or quick previews
• 720p – higher-quality output suitable for most publishing workflows

seed (optional)

Controls stochastic variation in generation:

• -1 or empty → random seed each run
• Any positive integer → reproducible results for the same inputs

(Exact field name may differ between Playground and API, but behavior is identical.)

🎯 Designed For

• Creators & influencers – Turn one performance into many identities without reshooting.
• Marketing & brands – Localize or personalize talking-head content while keeping the same body and scene.
• Film, TV & post-production – Rapid previs, mockups, and concept tests for head-replacement shots.
• Privacy & compliance – Replace real heads with synthetic or authorized identities while preserving situational context.

▶️ How to Use

1. Upload or paste the URL of the video to edit.
2. Upload a face/head reference image for the identity you want to swap in.
3. Select the output resolution (480p or 720p).
4. (Optional) Set a seed if you need reproducible results.
5. Click Run to generate the swapped video.
6. Review the result; if needed, adjust the reference portrait or seed and run again.

📌 Tips & Notes

• Use sharp, well-lit videos where the face is not heavily occluded or motion-blurred.
• For the reference portrait, keep expression and angle reasonably close to the target shot for the cleanest match.
• Avoid extreme mismatches in lighting (e.g., dark blue stage light in video vs. warm daylight portrait) unless you want a stylized look.
• Ensure you have the legal right and consent to use all uploaded videos and portraits.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
  "face_image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "resolution": "480p"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/video-head-swap" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters