Create Your First Project

This tutorial walks you through creating a complete annotation project in Mission Control.

What You’ll Learn

• How to create and configure a dataset
• How to set up an annotation project
• How to use the annotation tools
• How to review and export annotations

Prerequisites

• An Avala account with access to Mission Control
• Sample images to annotate (10-20 images recommended)

Step 1: Create a Dataset

1. Log in to Mission Control 
2. Click Datasets in the left sidebar
3. Click Add Dataset in the top right
4. Fill in the dataset details:
   • Name: “My First Dataset”
   • Description: Brief description of your data
   • Visibility: Private (or Public if sharing)
5. Click Create

Step 2: Upload Images

1. On your dataset page, click Add Items
2. Drag and drop your images into the upload zone
3. Alternatively, click to browse and select files
4. Wait for uploads to complete

Step 3: Create a Project

1. Navigate to Projects in the sidebar
2. Click Add Project
3. Configure your project:
   • Name: “Object Detection Project”
   • Dataset: Select your dataset
   • Task Type: Bounding Box
4. Click Next to configure labels

Step 4: Define Labels

Set up your label taxonomy:

1. Click Add Label
2. Enter label details:
   • Name: “car”
   • Color: Choose a color
3. Add more labels as needed (e.g., “pedestrian”, “bicycle”)
4. Click Create Project

Step 5: Start Annotating

1. Open your dataset → Items
2. Click on an image to open the annotation editor
3. Use the Box tool (press B) to draw bounding boxes
4. Draw a box around an object:
   • Click and drag from corner to corner
   • Release to complete the box
5. Select a label from the dropdown
6. Repeat for all objects in the image
7. Click Save or press Ctrl+S

Annotation Tips

• Tight boxes: Draw boxes that tightly fit the object
• Occluded objects: Include partially visible objects
• Consistent labeling: Use the same label for similar objects
• Keyboard shortcuts: Use shortcuts for faster annotation

Step 6: Review Annotations

1. Go to Projects → your project
2. Click Review to see pending annotations
3. For each annotation:
   • Accept: Annotation is correct
   • Reject: Annotation needs correction (add feedback)
4. Track progress in the project dashboard

Step 7: Export Data

1. Navigate to Exports
2. Click Create Export
3. Select:
   • Source: Your project
   • Format: COCO JSON (or your preferred format)
4. Click Create
5. Download when ready

Keyboard Shortcuts

Next Steps

• Learn about annotation tools in depth
• Set up quality control workflows
• Explore 3D annotation for point clouds