Depth Anything Video API Documentation

Depth Anything Video API Documentation

Playground

Try it on WaveSpeedAI!

Depth Anything Video estimates temporally consistent depth maps from video input, with stable depth across the whole clip and no flicker. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Depth Anything Video estimates dense, pixel-wise depth from an input video and returns a grayscale depth-map video. Depth remains consistent across the clip, helping reduce frame-to-frame flicker and brightness shifts while preserving the spatial structure of moving subjects, camera motion, and scene geometry.

The generated depth video follows the source resolution and duration and keeps the source audio track, with frame rate preserved up to 30 fps, making it suitable for depth-controlled video generation, VFX, compositing, AR, and 3D-aware editing workflows.


Why Choose This?

  • Temporally consistent depth
    Depth is normalized across the full video rather than independently frame by frame, helping maintain stable depth values as subjects and cameras move.

  • Stable through fast motion
    Depth is estimated with temporal context across neighboring frames, so structure holds up through fast camera moves and motion blur instead of changing from frame to frame.

  • Fine-grained scene structure
    Preserve silhouettes, hands, thin objects, foreground subjects, and background geometry.

  • Broad scene support
    Works across indoor, outdoor, live-action, and rendered footage without scene-specific tuning.

  • Long-video support
    Process up to 600 seconds of video in a single request.

  • Original-resolution output
    The generated depth video follows the input video resolution.

  • Frame-rate preservation
    Output follows the source frame rate up to 30 fps.


Parameters

ParameterRequiredDescription
videoYesInput video provided as a URL or upload. Up to 600 seconds are processed; longer inputs are processed up to the 10-minute mark.

How to Use

  1. Provide an input video — Upload a video or provide its URL.
  2. Submit — Run the depth-estimation task.
  3. Retrieve the result — Download the generated grayscale depth-map video.
  4. Use the output — Feed the depth video into depth-controlled generation, compositing, VFX, parallax, or other spatial workflows.

Pricing

Pricing is based on input video duration.

The rate is $0.005 per billed second.

Billed duration is rounded up to the next whole second, with a minimum billed duration of 3 seconds and a maximum billed duration of 600 seconds.

Input DurationBilled DurationCost
3s3s$0.015
10s10s$0.05
60s60s$0.30
600s600s$3.00

Best Use Cases

  • Depth-controlled video generation — Use the depth video as a structural control signal for generating new content while following the source motion and camera path.
  • Motion and camera replication — Extract depth from a reference clip to guide another video-generation workflow.
  • Cinematography and VFX — Apply depth-of-field, fog, volumetric lighting, and depth-aware effects in post-production.
  • Compositing — Use scene depth for masking, layering, and spatial placement.
  • AR occlusion — Help virtual objects pass correctly in front of or behind real-world subjects.
  • Motion graphics — Use depth as a displacement source for transitions and spatial animation.
  • 3D-aware editing — Apply depth-driven transformations while retaining scene structure over time.

Pro Tips

  • White represents surfaces closer to the camera, while black represents surfaces farther away.
  • Treat the generated result as relative depth, not real-world metric distance.
  • Use clips with clear foreground and background separation when strong depth structure is important.
  • Temporal consistency makes the output better suited to video-control workflows than independently estimating depth frame by frame.
  • Videos above 30 fps are returned at 30 fps while preserving the same duration.
  • The output keeps the source audio, so the depth video stays in sync when you lay it back over the original edit.
  • Use Depth Anything V3 Video when you need selectable grayscale or colorized depth visualizations.

Notes

  • video is required.
  • Up to 600 seconds of input video are processed.
  • Longer videos are processed only up to the first 600 seconds.
  • Output resolution follows the input video resolution.
  • Output frame rate follows the source up to 30 fps.
  • Inputs above 30 fps are returned at 30 fps with the same duration.
  • The source audio track is kept in the output video.
  • Depth is relative rather than metric.
  • Grayscale output uses lighter values for nearer surfaces and darker values for farther surfaces.

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"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/depth-anything/video" \
  -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

ParameterTypeRequiredDefaultRangeDescription
videostringYes-The URL of the input video to estimate depth for. Up to 10 minutes; longer videos are processed up to the 10-minute mark.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.statusstringTask status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses.
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.statusstringStatus: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.