> 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/instance-segmentation.md).

# Instance Segmentation

Trace the exact outline of each object in an image, then train and deploy your own instance segmentation model on Roboflow.

## What is instance segmentation?

Instance segmentation finds each object in an image and traces its exact outline. For each object, the model returns a class label, a confidence score, a box, and a mask (a polygon or pixel area) that covers only that object. Two touching objects of the same class get separate masks. Use it when you need the shape, size, or area of each object, not just its location.

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-3837aabcf19314a3c38f0171e4ee527c826671fe%2Fimg-task-instance-segmentation-wide.webp?alt=media" alt="Oranges on a conveyor, each traced with its own mask outline"><figcaption></figcaption></figure>

## Supported instance segmentation models

Roboflow supports these instance segmentation models. See the [Instance Segmentation models](https://docs.roboflow.com/models/supported-models.md#instance-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">Apache 2.0</a> (free for commercial use)</td><td>true</td><td>2026</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/roboflow-3.md">Roboflow 3.0</a></td><td><a href="https://roboflow.com/licensing">Roboflow Licensing</a></td><td>true</td><td>2023</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/sam3.md">SAM3</a></td><td><a href="https://docs.roboflow.com/models/model-licenses.md#other-models">Upstream license</a></td><td>true</td><td>2025</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/yolo11.md">YOLO11</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>2024</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/sam2.md">SAM2</a></td><td><a href="https://docs.roboflow.com/models/model-licenses.md#other-models">Upstream license</a></td><td>false</td><td>2024</td></tr><tr><td><a href="https://docs.roboflow.com/models/supported-models/yolov7.md">YOLOv7</a></td><td><a href="https://docs.roboflow.com/models/model-licenses.md">GPL-3.0</a></td><td>false</td><td>2022</td></tr></tbody></table>

## Fine-tuning an instance segmentation model

{% stepper %}
{% step %}

### Create an instance segmentation project

Create a [Project](https://docs.roboflow.com/datasets/create-and-upload/create-a-project) and pick "Instance 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-3ca9b7a4734d2129efed4ad5cec477b65d1b7cc6%2Fcreate-project-instance-segmentation.webp?alt=media" alt="Create project form with Instance Segmentation selected as the project type and a preview of an instance 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 them. The fastest way is [Auto Label](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/auto-label): a foundation model (ex: SAM 3, or Gemini or GPT paired with SAM 3) or one of your trained models outlines every object as a polygon or mask, and you review the results. You can also label in the [annotation editor](https://docs.roboflow.com/datasets/annotate/annotate/annotation-tools), either with AI help ([Smart Select](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/smart-select) outlines an object in one click, [Label Assist](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/model-assisted-labeling) suggests shapes from a model) or by hand with the Polygon tool.

<figure><img src="https://3448603185-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fcmo9mhfIjYrvFFy1U7dk%2Fuploads%2Fgit-blob-7ed16c839bebdab419502360b15eb5fc73e2bce9%2Fannotate-instance-segmentation.webp?alt=media" alt="Annotation editor on an instance segmentation image with polygons outlining fire and smoke, the class list on the left, and the labeling toolbar on the right"><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 an instance 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-20b1a8007a185a54d2577454cdc4c11119739c9b%2Ftrain-architectures-instance-segmentation.webp?alt=media" alt="Select Architecture step on the Train page listing RF-DETR (Recommended), YOLO26, SAM 3, Roboflow 3.0, and YOLOv11"><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

## Deploy an instance segmentation model

After training, deploy your instance 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.
