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

# Shapes and Objects

> Complete guide to all annotation types - bounding boxes, polygons, polylines, cuboids, points, ellipses, skeletons, masks, and tags

CVAT supports eight shape types and frame-level tags, providing comprehensive annotation capabilities for diverse computer vision tasks. Each annotation type is optimized for specific use cases and object characteristics.

## Annotation Types Overview

<CardGroup cols={3}>
  <Card title="Rectangle" icon="square">
    Axis-aligned bounding boxes
  </Card>

  <Card title="Polygon" icon="draw-polygon">
    Arbitrary closed shapes
  </Card>

  <Card title="Polyline" icon="wave-pulse">
    Open line segments
  </Card>

  <Card title="Points" icon="circle-dot">
    Individual point markers
  </Card>

  <Card title="Ellipse" icon="circle">
    Elliptical shapes
  </Card>

  <Card title="Cuboid" icon="cube">
    3D boxes (2D projection)
  </Card>

  <Card title="Skeleton" icon="person">
    Articulated keypoint structures
  </Card>

  <Card title="Mask" icon="paintbrush">
    Pixel-accurate segmentation
  </Card>

  <Card title="Tag" icon="tag">
    Frame-level labels
  </Card>
</CardGroup>

## Rectangle (Bounding Box)

### Description

Rectangles are axis-aligned bounding boxes defined by opposite corners. They're the fastest annotation type and are widely used for object detection.

### When to Use

* Object detection tasks
* When exact boundaries aren't critical
* Quick initial annotation
* Objects with roughly rectangular extent
* YOLO, Faster R-CNN, and similar detector training

### Drawing Methods

**Method 1: Classic (2 Points)**

1. Activate rectangle tool (`Shift+B`)
2. Click and drag from one corner to the opposite corner
3. Release to finish

**Method 2: Extreme Points (4 Points)**

1. Select "By 4 Points" in the tool options
2. Click the four extreme points:
   * Topmost point of the object
   * Bottommost point
   * Leftmost point
   * Rightmost point
3. CVAT fits a bounding box automatically

<Tip>
  Extreme points method is faster for objects at an angle or with irregular shapes, as you don't need to estimate the bounding box corners.
</Tip>

### Properties

* **Coordinates**: `(x1, y1, x2, y2)` where (x1, y1) is top-left and (x2, y2) is bottom-right
* **Rotation**: Can be rotated after creation (not axis-aligned anymore)
* **Attributes**: Supports all attribute types
* **In tracks**: Interpolates position and size

### Best Practices

<CardGroup cols={2}>
  <Card title="Tight Fit" icon="compress">
    Keep minimal margin around the object. The box should touch the object at the extreme points.
  </Card>

  <Card title="Include Full Object" icon="expand">
    Don't cut off any part of the object, even if partially occluded (mark as occluded instead).
  </Card>

  <Card title="Consistency" icon="equals">
    Maintain consistent margin across all instances of the same class.
  </Card>

  <Card title="Overlap Handling" icon="layer-group">
    Use Z-order to indicate which object is in front when boxes overlap.
  </Card>
</CardGroup>

***

## Polygon

### Description

Polygons are closed shapes defined by a sequence of vertices. They provide precise object boundaries for irregular shapes.

### When to Use

* Instance segmentation tasks
* Objects with irregular boundaries
* Precise boundary annotation required
* Semantic segmentation with discrete objects
* Mask R-CNN and similar model training

### Drawing

1. Activate polygon tool (`Shift+P`)
2. Click to place each vertex along the object boundary
3. Place vertices at corners and direction changes
4. Close the polygon:
   * Double-click the first vertex, or
   * Press `N`, or
   * Click the first point again

<Tip>
  **Vertex Placement Strategy**: Place more vertices where the boundary curves sharply, fewer on straight edges. Aim for smooth, accurate boundaries without excessive points.
</Tip>

### Editing

**Add Vertex**:

1. Select the polygon
2. Click on an edge where you want to insert a point
3. A new vertex is created

**Move Vertex**:

1. Select the polygon
2. Drag any vertex to reposition it

**Remove Vertex**:

1. Select the polygon
2. Right-click a vertex → "Delete point", or
3. Select vertex and press `Del`

**Context Menu**:

* Right-click edge → Add point
* Right-click vertex → Delete point

### Properties

* **Coordinates**: Array of `(x, y)` vertices
* **Minimum points**: 3 (triangle)
* **Attributes**: Supports all attribute types
* **In tracks**: Interpolates vertex positions (can morph shape)

### Best Practices

* **Zoom in** for pixel-accurate boundaries
* **Follow edges** precisely, especially for training segmentation models
* **Close gaps** carefully at occlusion boundaries
* **Optimize points**: Use enough for accuracy, not excessive
* **Clockwise/counterclockwise**: Maintain consistent direction for all objects

***

## Polyline

### Description

Polylines are open line segments (not closed). They represent linear features without interior regions.

### When to Use

* Lane markings
* Trajectories and paths
* Borders and boundaries
* Linear infrastructure (roads, power lines)
* Centerlines
* Motion paths

### Drawing

1. Activate polyline tool (`Shift+L`)
2. Click to place vertices along the line
3. Finish the line:
   * Press `N`, or
   * Double-click the last point

### Editing

Same as polygons:

* Add/remove vertices
* Drag vertices to reposition
* Extend or shorten the line

### Properties

* **Coordinates**: Array of `(x, y)` vertices (ordered)
* **Minimum points**: 2 (line segment)
* **Open shape**: No interior, just the path
* **In tracks**: Interpolates as a flexible path

### Best Practices

* Place vertices at **direction changes** and curves
* For lanes, annotate the **center line** unless specified otherwise
* Maintain **consistent direction** (e.g., always left-to-right)
* Extend to **frame boundaries** if the feature continues beyond the visible area

<Info>
  Polylines are especially useful for autonomous driving datasets where lane detection is critical.
</Info>

***

## Points

### Description

Points are individual point markers used for small objects, centers, or multiple instance counting.

### When to Use

* Very small objects (\< 5 pixels)
* Object centers or key points (single point per object)
* Counting (e.g., cells, people in crowds)
* Sparse annotation
* Click-based detection

### Drawing

1. Activate points tool (`Shift+.`)
2. Optionally set the number of points to place
3. Click to place each point
4. Press `N` after the last point (or auto-completes after specified count)

### Multiple Points in One Annotation

Points annotations can contain multiple points:

* Each click adds a point to the same annotation
* Useful for objects with multiple markers
* All points share the same label and attributes

### Properties

* **Coordinates**: Array of `(x, y)` positions
* **Minimum points**: 1
* **In tracks**: Interpolates point positions

### Best Practices

<CardGroup cols={2}>
  <Card title="Center Placement" icon="bullseye">
    When marking object centers, be consistent (geometric center, visual center, etc.)
  </Card>

  <Card title="Zoom In" icon="magnifying-glass-plus">
    Always zoom in for pixel-accurate point placement
  </Card>

  <Card title="Single vs Multiple" icon="layer-group">
    Use single-point annotations for discrete objects, multi-point for objects with multiple markers
  </Card>

  <Card title="Visibility" icon="eye">
    Points can be hard to see - increase point size in settings
  </Card>
</CardGroup>

***

## Ellipse

### Description

Ellipses are oval shapes defined by center and two radii. They provide a middle ground between rectangles and polygons.

### When to Use

* Circular or oval objects (balls, wheels, heads)
* When rotation matters
* Faster than polygons, more accurate than rectangles
* Objects with elliptical cross-sections

### Drawing

1. Activate ellipse tool
2. Click and drag to define center and initial size
3. Move mouse to adjust shape and rotation
4. Click to finish

### Editing

* **Drag center**: Move the ellipse
* **Drag edge handles**: Resize radii
* **Rotation handles**: Rotate the ellipse

### Properties

* **Coordinates**: Center `(cx, cy)`, radii `(rx, ry)`, and rotation angle
* **Shape**: Always elliptical
* **In tracks**: Interpolates position, size, and rotation

### Best Practices

* **Fit tightly** to the object boundary
* Use for **circular objects** where rectangles waste space
* Consider **rotation** for tilted objects
* **Alternative to polygons** when approximation is acceptable

***

## Cuboid

### Description

Cuboids are 3D boxes rendered in 2D as wire frames. They represent objects with depth in monocular images.

### When to Use

* Vehicles (cars, trucks, buses)
* Buildings and structures
* Furniture
* Packages and boxes
* Any object where 3D extent matters
* Monocular 3D object detection

### Drawing Methods

**Method 1: From Rectangle**

1. Activate cuboid tool
2. Draw the front face as a rectangle
3. Drag to extend the depth (rear face)
4. Adjust perspectives and edges

**Method 2: By 4 Points**

1. Select "By 4 Points" method
2. Click four corners that define the visible faces
3. CVAT constructs the 3D cuboid

### Structure

A cuboid consists of:

* **Front face**: 4 points
* **Rear face**: 4 points
* **Edges**: Connecting lines

Total: 8 vertices defining the 3D box.

### Editing

* **Drag faces**: Move front or rear face
* **Drag vertices**: Adjust individual corners
* **Drag edges**: Adjust perspective
* **Whole cuboid**: Drag center to move entire object

### Properties

* **Coordinates**: 8 vertices `(x, y)` for each corner
* **Projection**: Perspective projection of 3D box onto 2D image
* **Attributes**: Supports dimensions (width, height, length)
* **In tracks**: Interpolates all 8 vertices

### Best Practices

<AccordionGroup>
  <Accordion title="Perspective Consistency">
    Ensure the cuboid follows the image perspective. Rear face should be smaller if farther away.
  </Accordion>

  <Accordion title="Visible Faces">
    Adjust vertices to match visible edges. Not all edges may be visible depending on viewpoint.
  </Accordion>

  <Accordion title="Dimension Attributes">
    Use width/height/length attributes to record physical dimensions when known.
  </Accordion>

  <Accordion title="Orientation">
    Maintain consistent front/rear face assignment (e.g., front = vehicle front).
  </Accordion>
</AccordionGroup>

<Info>
  Cuboids are particularly valuable for autonomous driving datasets where 3D understanding is critical for planning and control.
</Info>

***

## Skeleton

### Description

Skeletons are articulated structures composed of keypoints (nodes) connected by edges. They represent objects with defined structural relationships.

### When to Use

* Human pose estimation
* Animal pose
* Articulated objects (robots, machinery)
* Hand keypoints
* Facial landmarks
* Any structured keypoint annotation

### Setup

Skeletons require label configuration with:

1. **Keypoints**: List of named points (e.g., "nose", "left\_eye", "right\_shoulder")
2. **Edges**: Connections between keypoints (e.g., "left\_shoulder" to "left\_elbow")

**Example Configuration** (COCO 17-keypoint pose):

```json theme={null}
{
  "name": "person",
  "type": "skeleton",
  "sublabels": [
    {"name": "nose"},
    {"name": "left_eye"},
    {"name": "right_eye"},
    {"name": "left_ear"},
    {"name": "right_ear"},
    {"name": "left_shoulder"},
    {"name": "right_shoulder"},
    // ... more keypoints
  ],
  "svg": "<skeleton edges definition>"
}
```

### Drawing

1. Activate skeleton tool
2. Select the skeleton label
3. Click to place each keypoint in order:
   * Follow the defined keypoint sequence
   * Place each point at the anatomical location
4. CVAT automatically draws edges between connected keypoints

<Tip>
  Zoom in when placing keypoints on small or distant people for accuracy. You can adjust keypoint positions after placement.
</Tip>

### Editing

* **Move keypoints**: Drag any keypoint to reposition
* **Visibility states**: Mark keypoints as visible/occluded/not visible
* **Add attributes**: Each keypoint can have visibility attributes
* **Edges**: Automatically update when keypoints move

### Properties

* **Structure**: N keypoints with M edges
* **Keypoint states**: Visible, occluded, outside frame
* **Coordinates**: `(x, y)` for each keypoint
* **In tracks**: Interpolates keypoint positions
* **Topology**: Fixed edge structure defined in label

### Common Skeleton Formats

**COCO Keypoints** (17 points):

* Full body pose with major joints
* Nose, eyes, ears, shoulders, elbows, wrists, hips, knees, ankles

**MPII** (16 points):

* Similar to COCO, slight variations

**Hand Keypoints** (21 points):

* Wrist + 4 fingers × 5 joints each

**Face Keypoints** (68+ points):

* Facial landmarks for expression analysis

### Best Practices

<CardGroup cols={2}>
  <Card title="Consistent Order" icon="list-ol">
    Always place keypoints in the defined order for consistency
  </Card>

  <Card title="Occluded Points" icon="eye-slash">
    Mark occluded keypoints appropriately, don't skip them
  </Card>

  <Card title="Symmetric Placement" icon="left-right">
    For symmetric objects (humans), maintain left/right consistency
  </Card>

  <Card title="Zoom for Accuracy" icon="magnifying-glass">
    Especially important for distant or small skeletons
  </Card>
</CardGroup>

***

## Mask

### Description

Masks are pixel-accurate binary segmentation masks. They represent objects at the highest level of detail.

### When to Use

* Pixel-perfect segmentation required
* Complex, irregular boundaries
* Objects with holes or intricate details
* Semantic segmentation
* Medical imaging
* High-precision applications

### Drawing

Masks use a specialized brush toolbox:

1. Activate mask tool
2. Select a label
3. The brush toolbox appears with tools:

**Brush Tool** (`Shift+1`):

* Paint to add mask pixels
* Adjustable size and shape (circle/square)
* Click and drag to paint

**Eraser Tool** (`Shift+2`):

* Remove mask pixels
* Same size/shape controls as brush

**Polygon Add** (`Shift+3`):

* Add a polygonal region to the mask
* Click vertices, close polygon
* Region is filled and added to mask

**Polygon Remove** (`Shift+4`):

* Remove a polygonal region from the mask
* Useful for quickly removing large areas

### Brush Settings

* **Size**: 1-100 pixels
* **Form**: Circle or square
* **Shortcuts**:
  * `[` - Decrease brush size
  * `]` - Increase brush size

<Tip>
  **Efficient Mask Workflow**:

  1. Use Polygon Add to roughly fill the object
  2. Use Brush to refine edges
  3. Use Eraser to remove overpainting
  4. Use Polygon Remove to cut out holes
</Tip>

### Editing

1. Select the mask annotation
2. The brush toolbox reappears
3. Use brush/eraser to modify
4. Click "Done" or press `N` to finish editing

### Properties

* **Storage**: Run-length encoded (RLE) for efficiency
* **Resolution**: Pixel-level
* **Binary**: Each pixel is either in or out of the mask
* **In tracks**: Each keyframe stores a full mask, interpolation is not applied (masks appear/disappear)

### Performance

Masks are computationally intensive:

* **Large masks**: May slow down the canvas
* **Many masks**: Consider using polygons if possible
* **Video masks**: Require keyframes on each annotated frame

### Best Practices

<AccordionGroup>
  <Accordion title="Use Polygon Tools First">
    Rough out the shape with Polygon Add, then refine with brush. This is much faster than painting everything.
  </Accordion>

  <Accordion title="Appropriate Brush Size">
    Use large brush for interiors, small brush for edges. Adjust frequently with `[` and `]`.
  </Accordion>

  <Accordion title="Zoom In for Edges">
    Zoom to at least 200% when painting boundary pixels for accuracy.
  </Accordion>

  <Accordion title="Save Frequently">
    Masks can be memory-intensive; save your work regularly to avoid loss.
  </Accordion>

  <Accordion title="Consider Auto-Annotation">
    Use SAM2 or other interactive tools to generate initial masks, then refine manually.
  </Accordion>
</AccordionGroup>

<Warning>
  In video annotation, masks do **not** interpolate between keyframes like other shape types. You must create a mask on each frame where the object is visible.
</Warning>

***

## Tag

### Description

Tags are frame-level labels without spatial extent. They classify entire frames or indicate frame properties.

### When to Use

* Image classification
* Scene classification
* Frame attributes (weather, time of day)
* Event detection (frame contains event X)
* Multi-label frame classification
* Metadata annotation

### Creating Tags

1. Click the Tag tool in the controls sidebar
2. Select a tag label
3. Click **Tag** to apply to current frame
4. The tag appears in the objects sidebar

<Info>
  Multiple tags can be applied to the same frame. Tags are independent of shape annotations.
</Info>

### Properties

* **No spatial extent**: Tags have no position or shape
* **Frame-specific**: Each tag applies to one frame
* **Attributes**: Tags support attributes
* **Label-based**: Each tag has a label (tag type)

### Tag Annotation Workspace

For rapid tag annotation, switch to the **Tag Annotation Workspace**:

1. Select workspace → "Tag Annotation"
2. Interface shows:
   * Large image/frame display
   * Tag buttons for quick application
   * Keyboard shortcuts for each tag
3. Navigate frames quickly with arrow keys
4. Apply tags with number keys or clicks

### Best Practices

* Use **consistent criteria** for tag application
* **Document tag definitions** in project guidelines
* Use **attributes** for tag variants (e.g., "weather" tag with "sunny"/"rainy" attribute)
* Leverage **keyboard shortcuts** for rapid annotation

***

## Object Types: Shape vs Track

All annotation types (except tags) can be created as either **shapes** or **tracks**:

### Shape

* Exists on a **single frame** only
* Used for **static images** or one-time appearances
* No temporal continuity
* **ObjectType**: `SHAPE`

### Track

* Exists across **multiple frames**
* Represents the same object over time
* Supports **interpolation** between keyframes
* Maintains a **single client ID** across frames
* **ObjectType**: `TRACK`

<Info>
  You can convert shapes to tracks and vice versa using the object context menu.
</Info>

### Track Keyframes

In tracks:

* **Keyframes**: Frames where you manually set the object's position/shape
* **Interpolated frames**: Frames between keyframes where CVAT automatically estimates position
* Mark a frame as keyframe with `K`

***

## Choosing the Right Annotation Type

<Tabs>
  <Tab title="By Task">
    | Task                  | Recommended Type |
    | --------------------- | ---------------- |
    | Object Detection      | Rectangle        |
    | Instance Segmentation | Polygon or Mask  |
    | Semantic Segmentation | Mask             |
    | Lane Detection        | Polyline         |
    | Pose Estimation       | Skeleton         |
    | Crowd Counting        | Points           |
    | Image Classification  | Tag              |
    | 3D Object Detection   | Cuboid           |
  </Tab>

  <Tab title="By Accuracy Needed">
    **Low Precision** (bounding box sufficient):

    * Rectangle
    * Ellipse

    **Medium Precision** (boundary approximation):

    * Polygon
    * Cuboid

    **High Precision** (pixel-perfect):

    * Mask
    * Skeleton (for keypoints)
  </Tab>

  <Tab title="By Speed">
    **Fastest**:

    * Rectangle (2-point)
    * Tag
    * Points (single)

    **Medium**:

    * Rectangle (4-point)
    * Ellipse
    * Polygon (few vertices)
    * Polyline

    **Slowest**:

    * Polygon (many vertices)
    * Mask (manual painting)
    * Skeleton (many keypoints)
    * Cuboid
  </Tab>
</Tabs>

## Export Formats

Different annotation types are supported by different export formats:

| Format          | Rectangle | Polygon | Polyline | Points | Mask | Skeleton | Cuboid | Tag |
| --------------- | --------- | ------- | -------- | ------ | ---- | -------- | ------ | --- |
| COCO            | ✓         | ✓       | ✗        | ✓      | ✓    | ✓        | ✗      | ✗   |
| YOLO            | ✓         | ✗       | ✗        | ✗      | ✗    | ✗        | ✗      | ✗   |
| Pascal VOC      | ✓         | ✗       | ✗        | ✗      | ✗    | ✗        | ✗      | ✗   |
| CVAT for images | ✓         | ✓       | ✓        | ✓      | ✓    | ✓        | ✓      | ✓   |
| CVAT for video  | ✓         | ✓       | ✓        | ✓      | ✓    | ✓        | ✓      | ✓   |
| Cityscapes      | ✓         | ✓       | ✗        | ✗      | ✗    | ✗        | ✗      | ✗   |
| KITTI           | ✓         | ✗       | ✗        | ✗      | ✗    | ✗        | ✓      | ✗   |

<Warning>
  Check your target export format before starting annotation to ensure it supports your chosen annotation types.
</Warning>

## Next Steps

<CardGroup cols={2}>
  <Card title="Manual Annotation" icon="hand-pointer" href="/annotation/manual-annotation">
    Learn workflows for creating these annotations
  </Card>

  <Card title="Advanced Tools" icon="wrench" href="/annotation/advanced-tools">
    Master interpolation and track editing
  </Card>

  <Card title="Auto-Annotation" icon="wand-magic-sparkles" href="/annotation/auto-annotation">
    Generate annotations automatically with AI
  </Card>

  <Card title="Export Datasets" icon="file-export" href="/datasets/import-export">
    Export your annotations in various formats
  </Card>
</CardGroup>