← All models
Generates depth maps from video using Video Depth Anything (CVPR 2025). Produces per-frame depth estimation with temporal consistency across frames. Supports 3 model sizes (Small, Base, Large), 5 colormaps including grayscale, side-by-side comparison with the original video, and raw depth export as .npz. Useful for 3D reconstruction, video effects, compositing, and scene understanding.
Example
Details
- Accepts
- video
Pricing
- Price
- 5 cr / second
Prices in credits (1 credit = $0.01).
Data schema
Input
| Field | Type | Description |
|---|---|---|
| model | stringenum: VDA-Small, VDA-Base, VDA-Large | Depth estimation model size. VDA-Large = best quality, VDA-Small = fastest. |
| colormap | stringenum: grayscale, turbo, inferno, magma, viridis | Colormap for depth visualization. 'turbo' (recommended) shows near=warm, far=cool. 'grayscale' for raw normalized depth. 'inferno'/'magma' for perceptually uniform. 'viridis' for colorblind-friendly. |
| video_urlrequired | string | URL of the input video to estimate depth for. |
| max_frames | — | Max frames to process (max 2400). None = up to 2400. |
| output_fps | — | Output video FPS. None = same as input. |
| resolution | stringenum: auto, 360p, 480p, 720p, 1080p | Output resolution. 'auto' preserves input (max 1080p). Options: 'auto', '360p', '480p', '720p', '1080p'. |
| side_by_side | boolean | Output original | depth comparison video. |
| include_raw_depths | boolean | Export raw float32 depths as .npz file with: 'depths' [N,H,W], 'min_depth', 'max_depth', 'fps', 'model', 'shape'. |
Output
| Field | Type | Description |
|---|---|---|
| video | — | Depth visualization video (MP4, H.264). |
| raw_depths | — | Raw depth values as .npz (if include_raw_depths=True). |