Class PerspectiveCamera

Camera that uses perspective projection.
This projection mode is designed to mimic the way the human eye sees

It is the most common projection mode used for rendering a 3D scene.

const camera = new THREE.PerspectiveCamera(45, width / height, 1, 1000);
scene.add(camera);

Hierarchy (view full)

Constructors

  • Creates a new PerspectiveCamera.

    Parameters

    • Optionalfov: number

      Camera frustum vertical field of view. Default 50.

    • Optionalaspect: number

      Camera frustum aspect ratio. Default 1.

    • Optionalnear: number

      Camera frustum near plane. Default 0.1.

    • Optionalfar: number

      Camera frustum far plane. Default 2000.

    Returns PerspectiveCamera

    Together these define the camera's viewing frustum.

Properties

DEFAULT_UP: Vector3

The default up direction for objects, also used as the default position for THREE.DirectionalLight | DirectionalLight,
THREE.HemisphereLight | HemisphereLight and THREE.Spotlight | Spotlight (which creates lights shining from the top down).

new THREE.Vector3( 0, 1, 0)

DEFAULT_MATRIX_AUTO_UPDATE: boolean

The default setting for matrixAutoUpdate for newly created Object3Ds.

true

DEFAULT_MATRIX_WORLD_AUTO_UPDATE: boolean

The default setting for matrixWorldAutoUpdate for newly created Object3Ds.

true

isPerspectiveCamera: true

Read-only flag to check if a given object is of type Camera.

This is a constant value

true

type: string

PerspectiveCamera

zoom: number

Gets or sets the zoom factor of the camera.

1

fov: number

Camera frustum vertical field of view, from bottom to top of view, in degrees.

Expects a Float

50

aspect: number

Camera frustum aspect ratio, usually the canvas width / canvas height.

Expects a Float

1, (square canvas).

near: number

Camera frustum near plane.

The valid range is greater than 0 and less than the current value of the .far plane.

0.1

far: number

Camera frustum far plane.

Must be greater than the current value of .near plane.

2000

focus: number

Object distance used for stereoscopy and depth-of-field effects.

This parameter does not influence the projection matrix unless a THREE.StereoCamera | StereoCamera is being used.

10

view: null | {
    enabled: boolean;
    fullWidth: number;
    fullHeight: number;
    offsetX: number;
    offsetY: number;
    width: number;
    height: number;
}

Frustum window specification or null.
This is set using the .setViewOffset method and cleared using .clearViewOffset.

null

filmGauge: number

Film size used for the larger axis.
This parameter does not influence the projection matrix unless .filmOffset is set to a nonzero value.

Expects a Float

35, millimeters.

filmOffset: number

Horizontal off-center offset in the same unit as .filmGauge.

Expects a Float

0

isCamera: true

Read-only flag to check if a given object is of type Camera.

This is a constant value

true

layers: Layers

The THREE.Layers | layers that the Camera is a member of.

Objects must share at least one layer with the Camera to be n when the camera's viewpoint is rendered.

new THREE.Layers()

matrixWorldInverse: Matrix4

This is the inverse of matrixWorld.

MatrixWorld contains the Matrix which has the world transform of the Camera .

THREE.Matrix4 | new THREE.Matrix4()

projectionMatrix: Matrix4

This is the matrix which contains the projection.

THREE.Matrix4 | new THREE.Matrix4()

projectionMatrixInverse: Matrix4

This is the inverse of projectionMatrix.

THREE.Matrix4 | new THREE.Matrix4()

coordinateSystem: CoordinateSystem
viewport?: Vector4
isObject3D: true

Flag to check if a given object is of type Object3D.

This is a constant value

true

id: number

Unique number for this Object3D instance.

Note that ids are assigned in chronological order: 1, 2, 3, ..., incrementing by one for each new object.
Expects a Integer

uuid: string

UUID of this object instance.

This gets automatically assigned and shouldn't be edited.

name: string

Optional name of the object

(doesn't need to be unique).

""

parent: null | Object3D<Object3DEventMap>

Object's parent in the scene graph.

An object can have at most one parent.

null

Array with object's children.

THREE.Object3DGroup | Group for info on manually grouping objects.

[]

This is used by the lookAt method, for example, to determine the orientation of the result.

Object3D.DEFAULT_UP - that is (0, 1, 0).

position: Vector3

Object's local position.

new THREE.Vector3() - that is (0, 0, 0).

rotation: Euler

Object's local rotation (Euler angles), in radians.

new THREE.Euler() - that is (0, 0, 0, Euler.DEFAULT_ORDER).

quaternion: Quaternion

Object's local rotation as a THREE.Quaternion | Quaternion.

new THREE.Quaternion() - that is (0, 0, 0, 1).

scale: Vector3

The object's local scale.

new THREE.Vector3( 1, 1, 1 )

modelViewMatrix: Matrix4

new THREE.Matrix4()

normalMatrix: Matrix3

new THREE.Matrix3()

matrix: Matrix4

The local transform matrix.

new THREE.Matrix4()

matrixWorld: Matrix4

The global transform of the object.

If the Object3D has no parent, then it's identical to the local transform THREE.Object3D.matrix | .matrix.

new THREE.Matrix4()

matrixAutoUpdate: boolean

When this is set, it calculates the matrix of position, (rotation or quaternion) and
scale every frame and also recalculates the matrixWorld property.

DEFAULT_MATRIX_AUTO_UPDATE - that is (true).

matrixWorldAutoUpdate: boolean

If set, then the renderer checks every frame if the object and its children need matrix updates.
When it isn't, then you have to maintain all matrices in the object and its children yourself.

DEFAULT_MATRIX_WORLD_AUTO_UPDATE - that is (true).

matrixWorldNeedsUpdate: boolean

When this is set, it calculates the matrixWorld in that frame and resets this property to false.

false

visible: boolean

Object gets rendered if true.

true

castShadow: boolean

Whether the object gets rendered into shadow map.

false

receiveShadow: boolean

Whether the material receives shadows.

false

frustumCulled: boolean

When this is set, it checks every frame if the object is in the frustum of the camera before rendering the object.
If set to false the object gets rendered every frame even if it is not in the frustum of the camera.

true

renderOrder: number

This value allows the default rendering order of scene graph
objects to be overridden although opaque and transparent objects remain sorted independently.

When this property is set for an instance of Group, all descendants objects will be sorted and rendered together.
Sorting is from lowest to highest renderOrder.

0

animations: AnimationClip[]

Array with object's animation clips.

[]

userData: Record<string, any>

An object that can be used to store custom data about the Object3D.

It should not hold references to functions as these will not be cloned.

{}

customDepthMaterial?: Material

Custom depth material to be used when rendering to the depth map.

Can only be used in context of meshes.
When shadow-casting with a THREE.DirectionalLight | DirectionalLight or THREE.SpotLight | SpotLight,
if you are modifying vertex positions in the vertex shader you must specify a customDepthMaterial for proper shadows.

undefined

customDistanceMaterial?: Material

Same as customDepthMaterial, but used with THREE.Object3DPointLight | PointLight.

undefined

Methods

  • Returns the focal length of the current .fov | fov in respect to .filmGauge.

    Returns number

  • Sets the FOV by focal length in respect to the current .filmGauge.

    Parameters

    • focalLength: number

      Expects a Float

    Returns void

    By default, the focal length is specified for a 35mm (full frame) camera.

  • Returns the current vertical field of view angle in degrees considering .zoom.

    Returns number

  • Returns the width of the image on the film

    Returns number

    If .aspect. is greater than or equal to one (landscape format), the result equals .filmGauge.

  • Returns the height of the image on the film

    Returns number

    If .aspect. is less than or equal to one (portrait format), the result equals .filmGauge.

  • Computes the 2D bounds of the camera's viewable rectangle at a given distance along the viewing direction.
    Sets minTarget and maxTarget to the coordinates of the lower-left and upper-right corners of the view rectangle.

    Parameters

    Returns void

  • Computes the width and height of the camera's viewable rectangle at a given distance along the viewing direction.
    Copies the result into the target Vector2, where x is width and y is height.

    Parameters

    Returns Vector2

  • Sets an offset in a larger frustum.

    Parameters

    • fullWidth: number

      Full width of multiview setup Expects a Float.

    • fullHeight: number

      Full height of multiview setup Expects a Float.

    • x: number

      Horizontal offset of subcamera Expects a Float.

    • y: number

      Vertical offset of subcamera Expects a Float.

    • width: number

      Width of subcamera Expects a Float.

    • height: number

      Height of subcamera Expects a Float.

    Returns void

    This is useful for multi-window or multi-monitor/multi-machine setups.

    For example, if you have 3x2 monitors and each monitor is 1920x1080 and
    the monitors are in grid like this

    ┌───┬───┬───┐
    ABC
    ├───┼───┼───┤
    DEF
    └───┴───┴───┘

    then for each monitor you would call it like this

      const w = 1920;
    const h = 1080;
    const fullWidth = w * 3;
    const fullHeight = h * 2;

    // Monitor - A
    camera.setViewOffset( fullWidth, fullHeight, w * 0, h * 0, w, h );
    // Monitor - B
    camera.setViewOffset( fullWidth, fullHeight, w * 1, h * 0, w, h );
    // Monitor - C
    camera.setViewOffset( fullWidth, fullHeight, w * 2, h * 0, w, h );
    // Monitor - D
    camera.setViewOffset( fullWidth, fullHeight, w * 0, h * 1, w, h );
    // Monitor - E
    camera.setViewOffset( fullWidth, fullHeight, w * 1, h * 1, w, h );
    // Monitor - F
    camera.setViewOffset( fullWidth, fullHeight, w * 2, h * 1, w, h );

    Note there is no reason monitors have to be the same size or in a grid.

  • Removes any offset set by the .setViewOffset method.

    Returns void

  • Updates the camera projection matrix

    Returns void

    Must be called after any change of parameters.

  • Parameters

    • focalLength: number
    • OptionalframeHeight: number

    Returns void

    Use .setFocalLength() and .filmGauge instead.

  • Returns a THREE.Vector3 | Vector3 representing the world space direction in which the Camera is looking.

    Parameters

    • target: Vector3

      The result will be copied into this Vector3.

    Returns Vector3

    Note: A Camera looks down its local, negative z-axis.

  • An optional callback that is executed immediately before a 3D object is rendered to a shadow map.

    Parameters

    Returns void

    This function is called with the following parameters: renderer, scene, camera, shadowCamera, geometry,
    depthMaterial, group.
    Please notice that this callback is only executed for renderable 3D objects. Meaning 3D objects which
    define their visual appearance with geometries and materials like instances of Mesh, Line,
    Points or Sprite. Instances of Object3D, Group or Bone are not renderable
    and thus this callback is not executed for such objects.

  • An optional callback that is executed immediately after a 3D object is rendered to a shadow map.

    Parameters

    Returns void

    This function is called with the following parameters: renderer, scene, camera, shadowCamera, geometry,
    depthMaterial, group.
    Please notice that this callback is only executed for renderable 3D objects. Meaning 3D objects which
    define their visual appearance with geometries and materials like instances of Mesh, Line,
    Points or Sprite. Instances of Object3D, Group or Bone are not renderable
    and thus this callback is not executed for such objects.

  • An optional callback that is executed immediately before a 3D object is rendered.

    Parameters

    Returns void

    This function is called with the following parameters: renderer, scene, camera, geometry, material, group.
    Please notice that this callback is only executed for renderable 3D objects. Meaning 3D objects which
    define their visual appearance with geometries and materials like instances of Mesh, Line,
    Points or Sprite. Instances of Object3D, Group or Bone are not renderable
    and thus this callback is not executed for such objects.

  • An optional callback that is executed immediately after a 3D object is rendered.

    Parameters

    Returns void

    This function is called with the following parameters: renderer, scene, camera, geometry, material, group.
    Please notice that this callback is only executed for renderable 3D objects. Meaning 3D objects which
    define their visual appearance with geometries and materials like instances of Mesh, Line,
    Points or Sprite. Instances of Object3D, Group or Bone are not renderable
    and thus this callback is not executed for such objects.

  • Applies the matrix transform to the object and updates the object's position, rotation and scale.

    Parameters

    Returns void

  • Applies the rotation represented by the quaternion to the object.

    Parameters

    Returns this

  • Calls THREE.Quaternion.setFromAxisAngle | setFromAxisAngle(axis, angle) on the .quaternion.

    Parameters

    • axis: Vector3

      A normalized vector in object space.

    • angle: number

      Angle in radians. Expects a Float

    Returns void

  • Calls THREE.Quaternion.setFromEuler | setFromEuler(euler) on the .quaternion.

    Parameters

    • euler: Euler

      Euler angle specifying rotation amount.

    Returns void

  • Calls THREE.Quaternion.setFromRotationMatrix | setFromRotationMatrix(m) on the .quaternion.

    Parameters

    • m: Matrix4

      Rotate the quaternion by the rotation component of the matrix.

    Returns void

    Note that this assumes that the upper 3x3 of m is a pure rotation matrix (i.e, unscaled).

  • Copy the given THREE.Quaternion | Quaternion into .quaternion.

    Parameters

    Returns void

  • Rotate an object along an axis in object space.

    Parameters

    • axis: Vector3

      A normalized vector in object space.

    • angle: number

      The angle in radians. Expects a Float

    Returns this

    The axis is assumed to be normalized.

  • Rotate an object along an axis in world space.

    Parameters

    • axis: Vector3

      A normalized vector in world space.

    • angle: number

      The angle in radians. Expects a Float

    Returns this

    The axis is assumed to be normalized
    Method Assumes no rotated parent.

  • Rotates the object around x axis in local space.

    Parameters

    • angle: number

    Returns this

  • Rotates the object around y axis in local space.

    Parameters

    • angle: number

    Returns this

  • Rotates the object around z axis in local space.

    Parameters

    • angle: number

    Returns this

  • Translate an object by distance along an axis in object space

    Parameters

    • axis: Vector3

      A normalized vector in object space.

    • distance: number

      The distance to translate. Expects a Float

    Returns this

    The axis is assumed to be normalized.

  • Translates object along x axis in object space by distance units.

    Parameters

    • distance: number

      Expects a Float

    Returns this

  • Translates object along y axis in object space by distance units.

    Parameters

    • distance: number

      Expects a Float

    Returns this

  • Translates object along z axis in object space by distance units.

    Parameters

    • distance: number

      Expects a Float

    Returns this

  • Converts the vector from this object's local space to world space.

    Parameters

    • vector: Vector3

      A vector representing a position in this object's local space.

    Returns Vector3

  • Converts the vector from world space to this object's local space.

    Parameters

    • vector: Vector3

      A vector representing a position in world space.

    Returns Vector3

  • Rotates the object to face a point in world space.

    Parameters

    • vector: Vector3

      A vector representing a position in world space to look at.

    Returns void

    This method does not support objects having non-uniformly-scaled parent(s).

  • Rotates the object to face a point in world space.

    Parameters

    • x: number

      Expects a Float

    • y: number

      Expects a Float

    • z: number

      Expects a Float

    Returns void

    This method does not support objects having non-uniformly-scaled parent(s).

  • Adds another Object3D as child of this Object3D.

    Parameters

    Returns this

    An arbitrary number of objects may be added
    Any current parent on an object passed in here will be removed, since an Object3D can have at most one parent.

    • attach
    • THREE.Group | Group for info on manually grouping objects.
  • Removes a Object3D as child of this Object3D.

    Parameters

    Returns this

    An arbitrary number of objects may be removed.

    THREE.Group | Group for info on manually grouping objects.

  • Removes this object from its current parent.

    Returns this

  • Removes all child objects.

    Returns this

  • Adds a Object3D as a child of this, while maintaining the object's world transform.

    Parameters

    Returns this

    Note: This method does not support scene graphs having non-uniformly-scaled nodes(s).

    add

  • Searches through an object and its children, starting with the object itself, and returns the first with a matching id.

    Parameters

    • id: number

      Unique number of the object instance. Expects a Integer

    Returns undefined | Object3D<Object3DEventMap>

    Note that ids are assigned in chronological order: 1, 2, 3, ..., incrementing by one for each new object.

    id

  • Searches through an object and its children, starting with the object itself, and returns the first with a matching name.

    Parameters

    • name: string

      String to match to the children's Object3D.name property.

    Returns undefined | Object3D<Object3DEventMap>

    Note that for most objects the name is an empty string by default
    You will have to set it manually to make use of this method.

  • Searches through an object and its children, starting with the object itself,
    and returns the first with a property that matches the value given.

    Parameters

    • name: string

      the property name to search for.

    • value: any

      value of the given property.

    Returns undefined | Object3D<Object3DEventMap>

  • Searches through an object and its children, starting with the object itself,
    and returns the first with a property that matches the value given.

    Parameters

    • name: string

      The property name to search for.

    • value: any

      Value of the given property.

    • OptionaloptionalTarget: Object3D<Object3DEventMap>[]

      target to set the result. Otherwise a new Array is instantiated. If set, you must clear
      this array prior to each call (i.e., array.length = 0;).

    Returns Object3D<Object3DEventMap>[]

  • Returns a vector representing the position of the object in world space.

    Parameters

    • target: Vector3

      The result will be copied into this Vector3.

    Returns Vector3

  • Returns a quaternion representing the rotation of the object in world space.

    Parameters

    • target: Quaternion

      The result will be copied into this Quaternion.

    Returns Quaternion

  • Returns a vector of the scaling factors applied to the object for each axis in world space.

    Parameters

    • target: Vector3

      The result will be copied into this Vector3.

    Returns Vector3

  • Abstract (empty) method to get intersections between a casted ray and this object

    Parameters

    Returns void

    Subclasses such as THREE.Mesh | Mesh, THREE.Line | Line, and THREE.Points | Points implement this method in order to use raycasting.

    THREE.Raycaster | Raycaster

    () => {}

  • Executes the callback on this object and all descendants.

    Parameters

    Returns void

    Note: Modifying the scene graph inside the callback is discouraged.

  • Like traverse, but the callback will only be executed for visible objects

    Parameters

    Returns void

    Descendants of invisible objects are not traversed.
    Note: Modifying the scene graph inside the callback is discouraged.

  • Executes the callback on all ancestors.

    Parameters

    Returns void

    Note: Modifying the scene graph inside the callback is discouraged.

  • Updates local transform.

    Returns void

  • Updates the global transform of the object.
    And will update the object descendants if .matrixWorldNeedsUpdate is set to true or if the force parameter is set to true.

    Parameters

    • Optionalforce: boolean

      A boolean that can be used to bypass .matrixWorldAutoUpdate, to recalculate the world matrix of the object and descendants on the current frame.
      Useful if you cannot wait for the renderer to update it on the next frame, assuming .matrixWorldAutoUpdate set to true.

    Returns void

  • Updates the global transform of the object.

    Parameters

    • updateParents: boolean

      Recursively updates global transform of ancestors.

    • updateChildren: boolean

      Recursively updates global transform of descendants.

    Returns void

  • Convert the object to three.js JSON Object/Scene format.

    Parameters

    • Optionalmeta: {
          geometries: any;
          materials: any;
          textures: any;
          images: any;
      }

      Object containing metadata such as materials, textures or images for the object.

      • geometries: any
      • materials: any
      • textures: any
      • images: any

    Returns any

  • Returns a clone of this object and optionally all descendants.

    Parameters

    • Optionalrecursive: boolean

      If true, descendants of the object are also cloned. Default true

    Returns this

  • Copies the given object into this object.

    Parameters

    • object: Object3D<Object3DEventMap>
    • Optionalrecursive: boolean

      If set to true, descendants of the object are copied next to the existing ones. If set to
      false, descendants are left unchanged. Default is true.

    Returns this

    Event listeners and user-defined callbacks (.onAfterRender and .onBeforeRender) are not copied.