Struct bevy::render::camera::Camera

pub struct Camera {
    pub viewport: Option<Viewport>,
    pub order: isize,
    pub is_active: bool,
    pub computed: ComputedCameraValues,
    pub target: RenderTarget,
    pub hdr: bool,
    pub output_mode: CameraOutputMode,
    pub msaa_writeback: bool,
    pub clear_color: ClearColorConfig,
}

Expand description

The defining Component for camera entities, storing information about how and what to render through this camera.

The Camera component is added to an entity to define the properties of the viewpoint from which rendering occurs. It defines the position of the view to render, the projection method to transform the 3D objects into a 2D image, as well as the render target into which that image is produced.

Adding a camera is typically done by adding a bundle, either the Camera2dBundle or the Camera3dBundle.

Fields§

§viewport: Option<Viewport>

If set, this camera will render to the given Viewport rectangle within the configured RenderTarget.

§order: isize

Cameras with a higher order are rendered later, and thus on top of lower order cameras.

§is_active: bool

If this is set to true, this camera will be rendered to its specified RenderTarget. If false, this camera will not be rendered.

§computed: ComputedCameraValues

Computed values for this camera, such as the projection matrix and the render target size.

§target: RenderTarget

The “target” that this camera will render to.

§hdr: bool

If this is set to true, the camera will use an intermediate “high dynamic range” render texture. This allows rendering with a wider range of lighting values.

§output_mode: CameraOutputMode

The CameraOutputMode for this camera.

§msaa_writeback: bool

If this is enabled, a previous camera exists that shares this camera’s render target, and this camera has MSAA enabled, then the previous camera’s outputs will be written to the intermediate multi-sampled render target textures for this camera. This enables cameras with MSAA enabled to “write their results on top” of previous camera results, and include them as a part of their render results. This is enabled by default to ensure cameras with MSAA enabled layer their results in the same way as cameras without MSAA enabled by default.

§clear_color: ClearColorConfig

The clear color operation to perform on the render target.

Implementations§

§

impl Camera

pub fn to_logical(&self, physical_size: UVec2) -> Option<Vec2>

Converts a physical size in this Camera to a logical size.

pub fn physical_viewport_rect(&self) -> Option<URect>

The rendered physical bounds URect of the camera. If the viewport field is set to Some, this will be the rect of that custom viewport. Otherwise it will default to the full physical rect of the current RenderTarget.

pub fn logical_viewport_rect(&self) -> Option<Rect>

The rendered logical bounds Rect of the camera. If the viewport field is set to Some, this will be the rect of that custom viewport. Otherwise it will default to the full logical rect of the current RenderTarget.

pub fn logical_viewport_size(&self) -> Option<Vec2>

The logical size of this camera’s viewport. If the viewport field is set to Some, this will be the size of that custom viewport. Otherwise it will default to the full logical size of the current RenderTarget. For logic that requires the full logical size of the RenderTarget, prefer Camera::logical_target_size.

Returns None if either:

the function is called just after the Camera is created, before camera_system is executed,
the RenderTarget isn’t correctly set:
- it references the PrimaryWindow when there is none,
- it references a Window entity that doesn’t exist or doesn’t actually have a Window component,
- it references an Image that doesn’t exist (invalid handle),
- it references a TextureView that doesn’t exist (invalid handle).

pub fn physical_viewport_size(&self) -> Option<UVec2>

The physical size of this camera’s viewport (in physical pixels). If the viewport field is set to Some, this will be the size of that custom viewport. Otherwise it will default to the full physical size of the current RenderTarget. For logic that requires the full physical size of the RenderTarget, prefer Camera::physical_target_size.

pub fn logical_target_size(&self) -> Option<Vec2>

The full logical size of this camera’s RenderTarget, ignoring custom viewport configuration. Note that if the viewport field is Some, this will not represent the size of the rendered area. For logic that requires the size of the actually rendered area, prefer Camera::logical_viewport_size.

pub fn physical_target_size(&self) -> Option<UVec2>

The full physical size of this camera’s RenderTarget (in physical pixels), ignoring custom viewport configuration. Note that if the viewport field is Some, this will not represent the size of the rendered area. For logic that requires the size of the actually rendered area, prefer Camera::physical_viewport_size.

pub fn target_scaling_factor(&self) -> Option<f32>

pub fn projection_matrix(&self) -> Mat4

The projection matrix computed using this camera’s CameraProjection.

pub fn world_to_viewport( &self, camera_transform: &GlobalTransform, world_position: Vec3 ) -> Option<Vec2>

Given a position in world space, use the camera to compute the viewport-space coordinates.

To get the coordinates in Normalized Device Coordinates, you should use world_to_ndc.

Returns None if any of these conditions occur:

The computed coordinates are beyond the near or far plane
The logical viewport size cannot be computed. See logical_viewport_size
The world coordinates cannot be mapped to the Normalized Device Coordinates. See world_to_ndc May also panic if glam_assert is enabled. See world_to_ndc.

Examples found in repository ?

examples/3d/blend_modes.rs (line 335)

fn example_control_system(
    mut materials: ResMut<Assets<StandardMaterial>>,
    controllable: Query<(&Handle<StandardMaterial>, &ExampleControls)>,
    mut camera: Query<(&mut Camera, &mut Transform, &GlobalTransform), With<Camera3d>>,
    mut labels: Query<(&mut Style, &ExampleLabel)>,
    mut display: Query<&mut Text, With<ExampleDisplay>>,
    labelled: Query<&GlobalTransform>,
    mut state: Local<ExampleState>,
    time: Res<Time>,
    input: Res<ButtonInput<KeyCode>>,
) {
    if input.pressed(KeyCode::ArrowUp) {
        state.alpha = (state.alpha + time.delta_seconds()).min(1.0);
    } else if input.pressed(KeyCode::ArrowDown) {
        state.alpha = (state.alpha - time.delta_seconds()).max(0.0);
    }

    if input.just_pressed(KeyCode::Space) {
        state.unlit = !state.unlit;
    }

    let randomize_colors = input.just_pressed(KeyCode::KeyC);

    for (material_handle, controls) in &controllable {
        let material = materials.get_mut(material_handle).unwrap();
        material.base_color.set_a(state.alpha);

        if controls.color && randomize_colors {
            material.base_color.set_r(random());
            material.base_color.set_g(random());
            material.base_color.set_b(random());
        }
        if controls.unlit {
            material.unlit = state.unlit;
        }
    }

    let (mut camera, mut camera_transform, camera_global_transform) = camera.single_mut();

    if input.just_pressed(KeyCode::KeyH) {
        camera.hdr = !camera.hdr;
    }

    let rotation = if input.pressed(KeyCode::ArrowLeft) {
        time.delta_seconds()
    } else if input.pressed(KeyCode::ArrowRight) {
        -time.delta_seconds()
    } else {
        0.0
    };

    camera_transform.rotate_around(Vec3::ZERO, Quat::from_rotation_y(rotation));

    for (mut style, label) in &mut labels {
        let world_position = labelled.get(label.entity).unwrap().translation() + Vec3::Y;

        let viewport_position = camera
            .world_to_viewport(camera_global_transform, world_position)
            .unwrap();

        style.top = Val::Px(viewport_position.y);
        style.left = Val::Px(viewport_position.x);
    }

    let mut display = display.single_mut();
    display.sections[0].value = format!(
        "  HDR: {}\nAlpha: {:.2}",
        if camera.hdr { "ON " } else { "OFF" },
        state.alpha
    );
}

pub fn viewport_to_world( &self, camera_transform: &GlobalTransform, viewport_position: Vec2 ) -> Option<Ray3d>

Returns a ray originating from the camera, that passes through everything beyond viewport_position.

The resulting ray starts on the near plane of the camera.

If the camera’s projection is orthographic the direction of the ray is always equal to camera_transform.forward().

To get the world space coordinates with Normalized Device Coordinates, you should use ndc_to_world.

Returns None if any of these conditions occur:

The logical viewport size cannot be computed. See logical_viewport_size
The near or far plane cannot be computed. This can happen if the camera_transform, the world_position, or the projection matrix defined by CameraProjection contain NAN. Panics if the projection matrix is null and glam_assert is enabled.

Examples found in repository ?

examples/3d/3d_viewport_to_world.rs (line 28)

fn draw_cursor(
    camera_query: Query<(&Camera, &GlobalTransform)>,
    ground_query: Query<&GlobalTransform, With<Ground>>,
    windows: Query<&Window>,
    mut gizmos: Gizmos,
) {
    let (camera, camera_transform) = camera_query.single();
    let ground = ground_query.single();

    let Some(cursor_position) = windows.single().cursor_position() else {
        return;
    };

    // Calculate a ray pointing from the camera into the world based on the cursor's position.
    let Some(ray) = camera.viewport_to_world(camera_transform, cursor_position) else {
        return;
    };

    // Calculate if and where the ray is hitting the ground plane.
    let Some(distance) = ray.intersect_plane(ground.translation(), Plane3d::new(ground.up()))
    else {
        return;
    };
    let point = ray.get_point(distance);

    // Draw a circle just above the ground plane at that position.
    gizmos.circle(
        point + ground.up() * 0.01,
        Direction3d::new_unchecked(ground.up()), // Up vector is already normalized.
        0.2,
        Color::WHITE,
    );
}

More examples

Hide additional examples

examples/3d/irradiance_volumes.rs (line 495)

fn handle_mouse_clicks(
    buttons: Res<ButtonInput<MouseButton>>,
    windows: Query<&Window, With<PrimaryWindow>>,
    cameras: Query<(&Camera, &GlobalTransform)>,
    mut main_objects: Query<&mut Transform, With<MainObject>>,
) {
    if !buttons.pressed(MouseButton::Left) {
        return;
    }
    let Some(mouse_position) = windows
        .iter()
        .next()
        .and_then(|window| window.cursor_position())
    else {
        return;
    };
    let Some((camera, camera_transform)) = cameras.iter().next() else {
        return;
    };

    // Figure out where the user clicked on the plane.
    let Some(ray) = camera.viewport_to_world(camera_transform, mouse_position) else {
        return;
    };
    let Some(ray_distance) = ray.intersect_plane(Vec3::ZERO, Plane3d::new(Vec3::Y)) else {
        return;
    };
    let plane_intersection = ray.origin + ray.direction.normalize() * ray_distance;

    // Move all the main objeccts.
    for mut transform in main_objects.iter_mut() {
        transform.translation = vec3(
            plane_intersection.x,
            transform.translation.y,
            plane_intersection.z,
        );
    }
}

pub fn viewport_to_world_2d( &self, camera_transform: &GlobalTransform, viewport_position: Vec2 ) -> Option<Vec2>

Returns a 2D world position computed from a position on this Camera’s viewport.

Useful for 2D cameras and other cameras with an orthographic projection pointing along the Z axis.

To get the world space coordinates with Normalized Device Coordinates, you should use ndc_to_world.

Returns None if any of these conditions occur:

The logical viewport size cannot be computed. See logical_viewport_size
The viewport position cannot be mapped to the world. See ndc_to_world May panic. See ndc_to_world.

Examples found in repository ?

examples/2d/2d_viewport_to_world.rs (line 25)

fn draw_cursor(
    camera_query: Query<(&Camera, &GlobalTransform)>,
    windows: Query<&Window>,
    mut gizmos: Gizmos,
) {
    let (camera, camera_transform) = camera_query.single();

    let Some(cursor_position) = windows.single().cursor_position() else {
        return;
    };

    // Calculate a world position based on the cursor's position.
    let Some(point) = camera.viewport_to_world_2d(camera_transform, cursor_position) else {
        return;
    };

    gizmos.circle_2d(point, 10., Color::WHITE);
}

pub fn world_to_ndc( &self, camera_transform: &GlobalTransform, world_position: Vec3 ) -> Option<Vec3>

Given a position in world space, use the camera’s viewport to compute the Normalized Device Coordinates.

When the position is within the viewport the values returned will be between -1.0 and 1.0 on the X and Y axes, and between 0.0 and 1.0 on the Z axis. To get the coordinates in the render target’s viewport dimensions, you should use world_to_viewport.

Returns None if the camera_transform, the world_position, or the projection matrix defined by CameraProjection contain NAN. Panics if the camera_transform contains NAN and the glam_assert feature is enabled.

pub fn ndc_to_world( &self, camera_transform: &GlobalTransform, ndc: Vec3 ) -> Option<Vec3>

Given a position in Normalized Device Coordinates, use the camera’s viewport to compute the world space position.

When the position is within the viewport the values returned will be between -1.0 and 1.0 on the X and Y axes, and between 0.0 and 1.0 on the Z axis. To get the world space coordinates with the viewport position, you should use world_to_viewport.

Returns None if the camera_transform, the world_position, or the projection matrix defined by CameraProjection contain NAN. Panics if the projection matrix is null and glam_assert is enabled.