---
title: "Create Image Recipes"
slug: "create-recipes"
updated: 2026-04-08T08:17:56Z
published: 2026-04-08T08:17:56Z
canonical: "docs.dataloop.ai/create-recipes"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.dataloop.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Create Image Recipes

In DDOE’s Annotation Platform, a **Recipe** is a comprehensive configuration that defines how data is annotated for a specific task, particularly for **image data**, including how images are displayed, what annotation tools are available, and how annotation outputs are structured.

You can create **image recipes** based on two types of annotation tasks:

- **Classic Image Tasks:** Use Classic Image Tasks to perform **image classification, tagging, and hierarchical annotations**. These tasks support **automation features** that help accelerate and scale labeling workflows.
- **Instance Segmentation Tasks:** Use Instance Segmentation Tasks to **precisely delineate individual object instances** within an image. These tasks map **each pixel to a specific object instance**, allowing you to distinguish between multiple objects of the same type within a single image.

This recipe supports both simple classification tasks and advanced pixel-level annotations, making it suitable for a wide range of vision-based AI applications.

**Common tasks include:**

- Image classification and tagging
- Object detection using bounding boxes or polygons
- Semantic and instance segmentation

These annotations are commonly used to train and validate computer vision models in domains such as retail, healthcare, manufacturing, and security.

## Start Creating Recipes for Classic Image Tasks

1. Open **Recipes** from the left-hand navigation panel.
2. Click **Create Recipe** to bring up the configuration dialog.

![](https://cdn.document360.io/53f32fe9-1937-4652-8526-90c1bc78d3f8/Images/Documentation/create_r_image.png)

1. Choose the **Image** type from the list.
2. Choose the **Classic Image Tasks** option from the list.
3. Select the annotation types from the list. The annotation tools will be auto-configured according to your annotation type selection:
  - Classification
  - Bulk Classification
  - Object Detection
  - Optical Character Recognition (OCR)
  - Semantic Segmentation
  - Other
4. Click **Create Recipe**. The recipe page is displayed and start configuring the following:
  - Labels
  - Attributes
  - Annotation tools
  - Instructions
  - Advanced Settings.

## Start Creating Recipes for Instance Segmentation Tasks

1. Open **Recipes** from the left-hand navigation panel.
2. Click **Create Recipe** to bring up the configuration dialog.

![](https://cdn.document360.io/53f32fe9-1937-4652-8526-90c1bc78d3f8/Images/Documentation/create_r_image.png)

1. Choose the **Image** type from the list.
2. Choose the **Instance Segmentation Task** option from the list. The annotation tools will be auto-configured according to your annotation type selection
3. Click **Create Recipe**. The recipe page is displayed and start configuring the following:
  - Labels
  - Attributes
  - Annotation tools
  - Instructions
  - Advanced Settings.

##