Video Cleaner API Documentation

Video Cleaner API Documentation

Playground

Video Cleaner removes people, vehicles and other moving subjects from a video and rebuilds the background behind them, producing a clean plate of the same shot with the camera motion intact. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Video Cleaner removes people, pedestrians, vehicles, and other moving subjects from a video, then rebuilds the background behind them. It preserves the source video’s camera movement, static scene structure, lighting, framing, and duration while creating a cleaner plate for editing, compositing, or restoration workflows.

Why Choose This?

• Moving-subject removal
  Remove people, crowds, cars, bikes, and other dynamic elements from video footage.

• Background reconstruction
  Rebuild the background where moving subjects were removed, including architecture, ground details, foliage, and static scene elements.

• Clean plate generation
  Create cleaner footage for compositing, advertising, restoration, or post-production workflows.

• Prompt-guided cleanup
  Use prompt to name extra objects that should also be removed.

• Multiple output resolutions
  Choose 480p, 720p, or 1080p while preserving the original aspect ratio.

• Seed support
  Set a seed when reproducible cleanup results are needed.

Parameters

How to Use

1. Upload a video — Provide the footage you want to clean.
2. Add a prompt optional — Name extra objects that should also disappear.
3. Choose target resolution — Select 480p, 720p, or 1080p.
4. Set seed optional — Use a fixed seed when reproducibility is needed.
5. Submit — Generate the cleaned video output.

Pricing

Pricing is based on input video duration and selected target_resolution.

Billed duration is rounded up to the next whole second, with a minimum billed duration of 3 seconds and a maximum billed duration of 120 seconds.

Example Costs

prompt and seed do not add separate charges.

Best Use Cases

• Clean plate creation — Remove moving subjects and generate a cleaner background plate.
• People and crowd removal — Remove pedestrians, crowds, or foreground people from footage.
• Vehicle removal — Remove cars, bikes, scooters, or other moving vehicles.
• Advertising workflows — Create clean plates for product compositing or campaign production.
• Post-production cleanup — Remove unwanted moving elements from short clips.
• Scene preparation — Prepare cleaner footage before relighting, colorization, restoration, or compositing.

Pro Tips

• Use footage where the background is visible around the moving subject.
• Large foreground subjects that cover most of the frame may leave too little background to reconstruct.
• Use prompt when a specific object remains after the first cleanup pass.
• Name objects clearly, such as bicycle, dog, car, or person in the background.
• Use clean source footage with stable motion when possible.
• Keep the same seed when you want to reproduce a result.
• For advertising plates, clean the clip first, then composite the product onto the result.

Limits and Performance

• Maximum processed video length per job: 120 seconds.
• Input beyond 120 seconds is not processed or billed.
• Processing time can vary by resolution and video content.
• Higher output resolutions generally take longer to process.

Notes

• video is required.
• target_resolution defaults to 720p.
• Aspect ratio is preserved.
• Billed duration is rounded up and clamped to 3–120 seconds.
• Very short videos are billed as 3 seconds.
• This endpoint is designed to remove moving subjects and rebuild visible background regions. It is not intended for changing the scene content or extending the video.

Related Models

• Video Restorer — Restore damaged footage by reducing compression artifacts or improving soft-focus video.
• Video Colorizer — Add or restore color in video footage for restoration and creative enhancement workflows.
• Video Cleaner — Clean up video noise, artifacts, and visual imperfections for clearer output.
• Video Relighter — Adjust or improve video lighting while preserving the original scene and motion.

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",
  "target_resolution": "720p"
}
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-cleaner" \
  -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