> For the complete documentation index, see [llms.txt](https://docs.roboflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.roboflow.com/models/supported-tasks/semantic-segmentation.md).

# Semantic Segmentation

Label every pixel in an image with a class to map areas and surfaces, then train and deploy your own semantic segmentation model on Roboflow.

## What is semantic segmentation?

Semantic segmentation gives every pixel in an image a class. The model returns a map the same size as the image, where each pixel holds a class (ex: "road", "water", "background"). It does not split objects of the same class apart: two touching cars become one "car" region. Use it for areas and surfaces rather than countable objects. To separate each object, use [Instance Segmentation](https://docs.roboflow.com/models/supported-tasks/instance-segmentation.md).

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-e98c6c79bff5d5a9f82abc134d42d8558d0cfa91%2Fimg-task-semantic-segmentation-wide.webp?alt=media" alt="Oranges, lemons, and limes on a table with every pixel colored by class: orange, lemon, lime, or background"><figcaption></figcaption></figure>

## Supported semantic segmentation models

Roboflow supports these semantic segmentation models. See the [Semantic Segmentation models](https://docs.roboflow.com/models/supported-models.md#semantic-segmentation) table for export and deployment support per model.

<table data-search="false"><thead><tr><th>Model</th><th>License</th><th data-type="checkbox">Trainable on Roboflow</th><th>Latest release</th></tr></thead><tbody><tr><td><a href="https://docs.roboflow.com/models/supported-models/rf-detr.md">RF-DETR</a></td><td><a href="https://docs.roboflow.com/models/model-licenses.md#platform-model-license">PML 1.0</a> (commercial use included on Roboflow cloud)</td><td>true</td><td>2026</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/yolo26.md">YOLO26</a></td><td><a href="https://docs.roboflow.com/models/model-licenses.md">AGPL-3.0</a> (commercial use needs a license, included on Roboflow cloud)</td><td>true</td><td>2026</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/roboflow-2.md">Roboflow 2.0</a></td><td><a href="https://roboflow.com/licensing">Roboflow Licensing</a></td><td>true</td><td>2022</td></tr></tbody></table>

## Fine-tuning a semantic segmentation model

{% stepper %}
{% step %}

### Create a semantic segmentation project

Create a [Project](https://docs.roboflow.com/datasets/create-and-upload/create-a-project) and pick "Semantic Segmentation" as the project type.

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-803b14d03919dc7a4fb71ea6c0dc84bd93244bdf%2Fcreate-project-semantic-segmentation.webp?alt=media" alt="Create project form with Semantic Segmentation selected as the project type and a preview of a semantic segmentation example"><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Upload and annotate images

[Upload images](https://docs.roboflow.com/datasets/create-and-upload/adding-data), then label each region in the [annotation editor](https://docs.roboflow.com/datasets/annotate/annotate/annotation-tools). Use [Smart Select](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/smart-select) to outline a region in one click or [Label Assist](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/model-assisted-labeling) to suggest masks from a model, or paint by hand with the Brush tool. Auto Label does not support semantic segmentation projects.

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-448ae80d4cc92a937e52a9e9af2844191b405580%2Fannotate-semantic-segmentation.webp?alt=media" alt="Annotation editor on a semantic segmentation image with road-lane and mid-lane masks painted over floor markings"><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Generate a dataset version

[Generate a dataset version](https://docs.roboflow.com/datasets/versions/dataset-versions/create-a-dataset-version): a frozen snapshot of your images with a train, valid, and test split, plus any preprocessing and augmentation.
{% endstep %}

{% step %}

### Train a semantic segmentation model

Click "Train", choose "Custom Training", and pick an architecture and size. See [Train a Model](https://docs.roboflow.com/models/train/train-a-model.md) for every option.

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-ac55c97c251fcc8208e5b936b2118db493ea66a1%2Ftrain-architectures-semantic-segmentation.webp?alt=media" alt="Select Architecture step on the Train page listing RF-DETR (Recommended), YOLO26, and Roboflow 2.0"><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

## Deploy a semantic segmentation model

After training, deploy your semantic segmentation model with the [Serverless Cloud API](https://docs.roboflow.com/deployment/roboflow-cloud/serverless-api), a [Dedicated Deployment](https://docs.roboflow.com/deployment/roboflow-cloud/dedicated-deployments), or on your own hardware with [Roboflow Inference](https://docs.roboflow.com/deployment/self-hosted/self-hosted). See [Choosing a Deployment](https://docs.roboflow.com/deployment/choosing-a-deployment) to compare the options.
