Header Logo

Docs

Start

Header Logo

Start

Header Logo

Docs

Start

All Models

WaveSpeed logo

Depth Anything V3 Video

Depth Anything V3 Video produces depth-map footage with clip-wide normalization and neighboring-frame context, supporting consistent depth through subject and camera movement.

Video editing

Based on verified input duration.

OfferingPrice (USD)
All listed options$0.005 / input second
  • Colormap: grayscale, inferno, turbo (same price).
  • Based on the video input; billing steps apply in this order:
  • Round input duration up to a whole second.
  • Cap billed duration at 600 seconds.
  • Bill at least 3 seconds.
  • Round the total charge up to the next $0.01.

Depth Anything V3 Video

Depth Anything V3 Video turns supplied footage into a depth-map video. It uses neighboring frames during estimation and normalizes depth across the clip, helping depth values remain consistent as the camera moves and subjects change position.

The output tracks spatial structure through time, including subject outlines and thin scene elements. Grayscale rendering supplies guidance for depth-conditioned video workflows; colorized rendering makes the estimated depth easier to inspect. The source footage provides the movement and scene arrangement represented in either view.

What it's good for

Compositors can use the depth pass to drive atmospheric effects that vary with distance, or to separate moving foreground subjects for focus and grading treatments. Temporal consistency helps those effects follow the action throughout a shot.

The depth video can also guide generation based on a reference clip's camera behavior. Other applications include depth-aware relighting, parallax treatments, and occlusion workflows that position virtual elements around filmed subjects.

Model ID

wavespeed-ai/depth-anything-v3-video

Output Type

Video

Pricing

$0.005 / input second

Provider

WaveSpeed

Hosting

Passthru

Depth Anything V3 Video

Depth Anything V3 Video turns supplied footage into a depth-map video. It uses neighboring frames during estimation and normalizes depth across the clip, helping depth values remain consistent as the camera moves and subjects change position.

The output tracks spatial structure through time, including subject outlines and thin scene elements. Grayscale rendering supplies guidance for depth-conditioned video workflows; colorized rendering makes the estimated depth easier to inspect. The source footage provides the movement and scene arrangement represented in either view.

What it's good for

Compositors can use the depth pass to drive atmospheric effects that vary with distance, or to separate moving foreground subjects for focus and grading treatments. Temporal consistency helps those effects follow the action throughout a shot.

The depth video can also guide generation based on a reference clip's camera behavior. Other applications include depth-aware relighting, parallax treatments, and occlusion workflows that position virtual elements around filmed subjects.