Source: https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/

# Preprocessors

A preprocessor takes the image (or video) you've loaded into the ControlNet panel and converts it into the _kind_ of cue your chosen ControlNet model expects: a depth map, a pose skeleton, an outline drawing, a segmentation mask, or a set of face landmarks. The preprocessor row sits just to the left of your reference image in the ControlNet panel.

You can preview each preprocessor's output before generating — useful for confirming the cue is what you wanted before you spend Spark on a render.

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#available-preprocessors)Available preprocessors

| Preprocessor | What it produces | Pair with |
| --- | --- | --- |
| **Face Capture** | Facial landmark map | OpenPose CN model |
| **Pose Capture** | OpenPose-style joint skeleton | OpenPose CN model |
| **Sketch / Outline** | Anime-style outline drawing | Canny, LineArt, LineArt Anime, M-LSD, Scribble, SoftEdge, Segmentation, Depth, Normal Bae |
| **Depth Map** | MiDaS-based depth estimation | Depth CN model |
| **Segmentation** | Class-labeled region map (U2Net / IS-Net) | Segmentation CN model |
| **RMBG (Background Removal)** | Subject isolated from background | Segmented Subject / Segmented Background CN |
| **Invert** | Inverts colors (white↔black) | Use after Sketch / Outline for line drawings |

Multiple preprocessors can be chained — for example, **Sketch + Invert** turns a photo into white lines on a black background, which is what Canny / LineArt models expect.

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#on-device-preprocessor-models)On-device preprocessor models

When you run on-device, the preprocessing step itself runs locally on your Mac. Sogni Studio lazy-loads these small models the first time you use a given preprocessor:

| Model | Size | Used for |
| --- | --- | --- |
| `anime2sketch` | 200 MB | Sketch / Outline preprocessor |
| `midas_small` | 61 MB | Depth Map preprocessor |
| `is_net` | 163 MB | IS-Net segmentation |
| `u2_net` | 165 MB | U2Net segmentation |
| `u2_netP` | 4 MB | Lightweight segmentation variant |
| `RMBG` | 165 MB | Background removal |

If you only ever render on Supernet, you don't need any of these — preprocessing runs on the worker.

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#face-capture)Face Capture

Locates all faces in the reference, detects landmarks, and emits a facial landmark map. Pair with the **OpenPose** ControlNet model to transfer expressions and head orientation. Works with one or more faces and can run together with Pose Capture for full body + face control.

![Face Capture preprocessor icon](https://docs.sogni.ai/assets/remote/bbfd50114ad9-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FH2SFtu.jpg)

| Reference image | Face Capture output |
| --- | --- |
| ![Reference portrait before Face Capture](https://docs.sogni.ai/assets/remote/bf62e21a09c2-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FUP9LtK.png) | ![Facial landmark map produced by Face Capture](https://docs.sogni.ai/assets/remote/4d1cc4386403-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FF5JHyN.png) |

![Example generations using Face Capture with OpenPose](https://docs.sogni.ai/assets/remote/6c2ab871b17c-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FiBNCLP.png)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#pose-capture)Pose Capture

Detects human subjects and extracts an OpenPose bone-joint skeleton. Pair with the OpenPose ControlNet model to lock the pose of a generated subject. Combine with Face Capture to lock pose **and** expression in a single generation.

![Pose Capture preprocessor icon](https://docs.sogni.ai/assets/remote/f1b8f7fc7006-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FfCat3H.jpg)

| Reference image | Pose Capture output |
| --- | --- |
| ![Reference image before Pose Capture](https://docs.sogni.ai/assets/remote/8d2035dd0c6f-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2Fl78KGS.png) | ![OpenPose skeleton produced by Pose Capture](https://docs.sogni.ai/assets/remote/b3f46e335ada-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FFnIMMr.png) |

![Example generations using Pose Capture with OpenPose](https://docs.sogni.ai/assets/remote/f42967d16160-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FbHWP0G.png)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#sketch-outline)Sketch / Outline

Converts a reference into a sketch-style line drawing. For best results, follow with **Invert** so you get white lines on a black background — what Canny, LineArt, and Scribble models are trained on.

Sketch works well with: **Depth, LineArt, LineArt Anime, M-LSD, Normal Bae, Scribble, Segmentation, SoftEdge.**

![Sketch and Outline preprocessor icon](https://docs.sogni.ai/assets/remote/56231d756f98-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2Fh3zlTe.jpg)

| Reference image | Sketch / Outline output |
| --- | --- |
| ![Reference image before Sketch processing](https://docs.sogni.ai/assets/remote/ab2c2db78236-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FjCGrZd.png) | ![Line drawing produced by Sketch processing](https://docs.sogni.ai/assets/remote/12f2bcf40584-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FyQ6eI6.png) |

![Applying Sketch and Invert to prepare ControlNet line art](https://docs.sogni.ai/assets/remote/3b1b24cc59fe-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2F7RvJzc.gif)

![Example generations using Sketch with the M-LSD ControlNet model](https://docs.sogni.ai/assets/remote/3e8b2ebc74ca-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FkeHKlB.png)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#depth-map)Depth Map

Runs MiDaS depth estimation locally (or its equivalent on Supernet) and produces a grayscale depth image. Pair with the **Depth** ControlNet model. Best for compositions where 3D structure matters — interiors, landscapes, portraits with strong foreground/background separation.

![Depth Map preprocessor icon](https://docs.sogni.ai/assets/remote/7003bd3862d2-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2F6HXGc5.jpg)

![Generating a depth map from a reference image](https://docs.sogni.ai/assets/remote/46c8ffe02416-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FQRnsfN.gif)

![Example generations using the Depth Map preprocessor and Depth ControlNet model](https://docs.sogni.ai/assets/remote/e4a9d909bc98-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FQyZuyl.png)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#segmentation)Segmentation

Splits the reference into labeled regions — sky, person, vehicle, building, etc. — using U2Net or IS-Net under the hood. Pair with the **Segmentation** ControlNet model to lock spatial layout while changing the content of each region via prompt.

![Segmentation preprocessor icon](https://docs.sogni.ai/assets/remote/2889ebaaa666-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FJ0Y15D.jpg)

![Generating a segmentation map and selecting subject or background masking](https://docs.sogni.ai/assets/remote/38b29572d925-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FxYav1B.gif)

### [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#subject-only-mask)Subject-only mask

![A subject-only segmentation mask](https://docs.sogni.ai/assets/remote/07c66f67ed64-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FFcNboU.png)

![Example generations using a subject-only mask](https://docs.sogni.ai/assets/remote/de6d34261eec-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2Fu45cxX.png)

### [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#background-only-mask)Background-only mask

![A background-only segmentation mask](https://docs.sogni.ai/assets/remote/1d122b762ef1-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2Fybt6ae.png)

![Example generations using a background-only mask](https://docs.sogni.ai/assets/remote/54c4f74885cb-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2F7DgTRk.png)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#rmbg-background-removal)RMBG (Background Removal)

Isolates the foreground subject from the background. Used to feed either:

-   **Segmented Subject** ControlNet — render the subject as-is, regenerate the background.
-   **Segmented Background** ControlNet — keep the background, regenerate the subject.

Useful for character-on-new-environment shots and background swaps.

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#invert)Invert

Not a real preprocessor — flips the colors of whatever's currently shown. Apply after Sketch / Outline when working with Canny, LineArt, or Scribble models.

![Invert Colors preprocessor icon](https://docs.sogni.ai/assets/remote/7f9cb9b7ca59-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2FvOcPyC.jpg)

![Inverting a preprocessed ControlNet image](https://docs.sogni.ai/assets/remote/a9be41b28c71-spaces-2FjGm8Y44m4zCkKjnnr9Na-2Fuploads-2F5GW9PL.gif)

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#tips)Tips

-   **Preview before you render.** The preprocessor preview is the cue that ControlNet actually sees. If the depth map looks flat or the pose skeleton missed limbs, the render will reflect that — fix at the preprocessing stage.
-   **Crop the reference first.** If your reference has multiple subjects but you only want one, crop down so preprocessors lock onto the target.
-   **Lower frame rates for video.** Video ControlNet preprocessing scales linearly with frame count — drop to 12 or 16 FPS if you only need motion direction, not every micro-movement.

## [#](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/preprocessors/#see-also)See also

[ControlNet— using cues to steer generation.](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/)[ControlNet Models— Canny, LineArt, Depth, OpenPose, Segmentation, InPaint, and more.](https://docs.sogni.ai/sogni-create/studio/creating-images/generation-settings/controlnet/controlnet-models/)