Layer Image

The Layer Image Image Generator outputs the image of another layer of the composition. It does not make a copy: it refers to that layer, so any change made to the original, its animation, its Modifiers or its content, is instantly reflected wherever the Layer Image is used.

Use it for:

• Live duplicates of a layer: reflections, shadows, echoes and trails that always follow the original
• Applying Modifiers to everything below it in the layer stack
• Reusing the original image of a layer inside its own Modifier stack
• Rendering a 3D part of the composition again, with other settings

Quick Navigation

• Creating and Connecting a Layer Image
• Choosing the Layer to Copy
• Choosing What to Copy
• Applying Modifiers to the Layers Below
• Reusing the Original Image of a Layer: This Layer
• Time Offsets and Echoes: Follow Time
• Parts Outside the Frame: Adjust To Source Format
• Referencing 3D Layers
• Controls

---

Creating and Connecting a Layer Image

Like any Image Generator, the Layer Image can be connected to any image parameter, for example:

• The Source of a layer, to display the copy
• The Mask of a layer, to use the shape of another layer as a mask
• The image input of a Modifier, for example to composite another layer, or the layer itself, inside a Modifier stack (see This Layer)

The quickest way to create a layer with a Layer Image as its source is described in Adding Layers:

• Click the + button at the top of the layer stack and choose Layer Image from the list of Generators
• Place the mouse cursor over the Viewer and open the Quick Access Menu, then pick Layer Image:
  • Press Q when no layer is selected
  • Press Shift+Q when a layer is selected (in this case, Q opens the menu dedicated to Modifiers)

To use it with any other image parameter, such as a Mask, click the parameter's slot and choose Layer Image from the list of Generators on the right.

By default, a new Layer Image copies the layer directly below it, blended with everything underneath (Layer Source set to Layer Below, Which set to Post Blend). The copy is drawn exactly where the original is, so it hides it: move the Layer Image layer with its Transform to see both.

Choosing the Layer to Copy

The Layer Source menu decides which layer the Layer Image refers to:

• Layer Below (the default): the layer directly below the one that hosts the Layer Image. Moving the layers in the stack changes the reference.
• Layer Above: the layer directly above.
• Linked Layer: a specific layer, wherever it is in the stack. A Layer field appears: drag and drop a layer from the Timeline onto it.
• This Layer: the layer that hosts the Layer Image itself. It is meant to be used inside the Modifier stack or the Mask of that layer, as described in This Layer.

Choosing What to Copy

A layer is built in several steps: its Source produces an image, its Transform places it, its Modifiers process it, and finally it is blended over the layers below. The Which menu decides at which step the Layer Image takes its copy.

• Source: the image of the layer's Source only, without its Transform and its Modifiers.
• Source With Transform: the Source placed by the layer's Transform, without its Modifiers.
• Source With Modifiers: the Source with both its Transform and its Modifiers, in other words the layer exactly as you see it, but alone.
• Post Blend (the default): the layer blended over everything below it in the stack. See Applying Modifiers to the Layers Below.
• 3D Render: only available when the Layer Image refers to 3D layers. See Referencing 3D Layers.

Applying Modifiers to the Layers Below

With Which set to Post Blend and Layer Source set to Layer Below, the Layer Image contains the whole composition as it looks below it. Any Modifier added to the Layer Image layer then processes all the layers below at once, and a Mask on that layer limits the effect to a part of the frame.

For this kind of global correction, Autograph also offers dedicated Adjustment Layers, which are simpler to set up. Use the Layer Image when you need the result of the layers below as an image, for example in the slot of a Modifier or in a Mask.

Reusing the Original Image of a Layer: This Layer

With Layer Source set to This Layer, the Layer Image brings back the original Source (or Mask) of the layer that hosts it. Place it in the image slot of a Modifier, and you can reinject the untouched original anywhere in the Modifier stack of that layer, after other Modifiers have changed it.

In the following example, a Polygon is the Source of the layer:

1. Edge Extend, Blur and Fill Modifiers turn the blue hexagon into a soft green glow.
2. A Merge Modifier is added at the end of the stack, with its Source Mode set to Foreground: it composites another image on top of the current state of the stack.
3. A Layer Image set to This Layer is connected to the Foreground slot of the Merge. The original blue hexagon is composited back over the glow.

Everything stays dynamic: change the size or the number of sides of the Polygon, and both the glow and the reinjected hexagon follow.

Time Offsets and Echoes: Follow Time

Every layer has its own local time, which can be shifted with its Time Offset. The Follow Time option decides at which time the copied layer is evaluated:

• On (the default): the copied layer is evaluated at its own time, so the copy always matches the original, whatever the Time Offset of the Layer Image layer.
• Off: the copied layer is evaluated at the time of the Layer Image layer. Shift that layer in time, and the copy lags behind the original, or gets ahead of it.

Turning Follow Time off is the key to echo and trail effects. In the following example, five Layer Image layers are stacked below an animated shape. Each one copies the layer just above it (Layer Source set to Layer Above), with a Time Offset of -0.1 second and a lower opacity, so each copy is one step further behind.

Parts Outside the Frame: Adjust To Source Format

In Autograph, an image is not limited to the Format of the composition: a layer partly moved outside the frame still exists outside of it, even if you do not see that part. The Adjust To Source Format option decides whether the Layer Image keeps those parts:

• Off (the default): the copy includes everything, even the parts of the original that lie outside the frame. When the copy is moved back into the frame, these parts become visible.
• On: the original is cropped by the Format before being copied. The copy is cut exactly where the original leaves the frame.

Referencing 3D Layers

In Autograph, the composition itself is the 3D scene: meshes, lights and cameras are layers of the composition. A 2D layer placed between 3D layers splits the scene in two sub-scenes, one below it and one above it.

When a Layer Image refers to 3D layers, for example with Layer Source set to Layer Below, it refers to the whole sub-scene: all the 3D layers below it, up to the previous 2D layer. The Which menu then offers a 3D Render entry, and new options appear:

• Use Existing 3D Render: when on (the default), the color image already rendered for the sub-scene is reused. When off, the Layer Image renders the sub-scene again, with its own settings.
• Renderer: the render engine used for the new render, Filament (the default) or Storm.
• AoV: the pass to render, Color or Depth. A Depth pass is useful to drive depth-based effects.
• Output Format and Camera: the size of the new render, and the camera it is rendered from (by default, the top-most active camera).
• Filament Settings: the quality options of the Filament renderer, such as ambient occlusion, depth of field, fog, anti-aliasing, shadows and reflections. They are listed in the Controls table below.

---

Controls

Parameter / Script Name	Type	Default	Function
Enabled / enabled	Boolean	On	Turns the Generator on or off. When off, the Generator produces an empty image.
Layer Source / layerSource	Choice	Layer Below	Which layer is copied. See Choosing the Layer to Copy. - Layer Below: The layer directly below. - Layer Above: The layer directly above. - Linked Layer: The layer set in the Layer field. - This Layer: The layer that hosts the Layer Image, for use in its own Modifier stack or Mask.
Layer / layer	Link	-	The layer to copy when Layer Source is set to Linked Layer. Drag and drop a layer from the Timeline onto this field.
Follow Time / followTime	Boolean	On	When on, the copied layer is evaluated at its own time, so the copy matches the original. When off, it is evaluated at the time of the layer that hosts the Layer Image, so a Time Offset on that layer delays or advances the copy. See Follow Time.
Which / which	Choice	Post Blend	At which step of the copied layer the image is taken. The available entries depend on the context. See Choosing What to Copy. - Source: The Source only, without Transform and Modifiers. - Source With Transform: The Source with the layer's Transform. - Source With Modifiers: The Source with the layer's Transform and Modifiers. - 3D Render: Only when 3D layers are copied: the render of the sub-scene. - Post Blend: The layer blended over everything below it.
Use Existing 3D Render / useExisting3DRender	Boolean	On	When checked, the color image already rendered for the referenced 3D sub-scene is reused. When unchecked, the sub-scene is rendered again with the settings below, for example to render a Depth pass or another format.
Renderer / render_delegate	Choice	Filament	The render engine used when Use Existing 3D Render is off. - Storm - Filament
AoV / aov	Choice	Color	The pass rendered when Use Existing 3D Render is off. - Color - Depth
Output Format / output_format_mode	Choice	Composition Format	The size of the new 3D render. - Preset: Uses the Format preset. - Composition Format: Uses the format of the composition. - Custom: Uses Width, Height and Pixel Aspect Ratio.
Format / formatPreset	Choice	HD (1920x1080)	- PC VIDEO (640x480) - NTSC (720x486:0.91) - PAL (720x576:1.09) - NTSC 16:9 (720x486:1.21) - PAL 16:9 (720x576:1.46) - QHD (960x540) - LHD (1280x720) - HD (1920x1080) - VHD (1080x1920) - UHD 4K (3840x2160) - VUHD 4K (2160x3840) - UHD 8K (7680x4320) - Super35 1K FULL AP (1024x778) - Cinemascope 1K (914x778:2.00) - Super35 2K FULL AP (2048x1556) - Cinemascope 2K (1828x1556:2.00) - DCP 2K (2048x1080) - Super35 4K FULL AP (4096x3112) - DCP 4K (4096x2160) - Square 256 (256x256) - Square 512 (512x512) - Square 1K (1024x1024) - Square 2K (2048x2048)
Width / width	Float	1920	Width of the new 3D render, in pixels, when Output Format is set to Custom.
Height / height	Float	1080	Height of the new 3D render, in pixels, when Output Format is set to Custom.
Pixel Aspect Ratio / customAspect	Float	1	Pixel aspect ratio of the new 3D render, when Output Format is set to Custom.
Camera / camera	Choice	Top-Most Active	The camera used for the new 3D render. By default, the top-most active camera of the composition.
Filament Settings / FilamentSettings	Category	-	Quality options of the Filament renderer, used when Use Existing 3D Render is off.
Ambient Occlusion / group:Ambient Occlusion	Category	-
Ambient Occlusion / filament_ao_type	Choice	None	The ambient occlusion technique, or None to turn ambient occlusion off. - None - SAO - GTAO
Bent Normals / filament_ao_bent_normals	Boolean	Off
Min. Horizon Angle / filament_min_horiz_angle	Float	0	Minimum angle to consider
Quality / filament_ao_quality	Choice	Ultra	- Low - Medium - High - Ultra
Radius / filament_ao_radius	Float	0.3	Ambient Occlusion radius in meters, between 0 and ~10.
Intensity / filament_ao_intensity	Float	1	Strength of the Ambient Occlusion effect
Bias / filament_ao_bias	Float	0.0005	Self-occlusion bias in meters. Use to avoid self-occlusion. Between 0 and a few mm
Bilateral threshold / filament_ao_bilateral_threshold	Float	0.05	Depth distance that constitute an edge for filtering
Power / filament_ao_power	Float	1	Controls ambient occlusion's contrast. Must be positive.
Upsampling / filament_ao_upsampling	Choice	Ultra	- Low - Medium - High - Ultra
Low pass filter / filament_ao_low_pass_filter	Choice	Ultra	- Low - Medium - High - Ultra
SSCT / filament_ssct_enabled	Boolean	Off	Enable or disable screen space cone tracing for ambient shadows of dominant lights
Light Cone Radius / filament_ssct_light_cone_radius	Float	1
SSCT Shadow Distance / filament_ssct_shadow_distance	Float	0.3	How far shadows can be cast
SSCT Contact Distance / filament_ssct_contact_distance	Float	1	Max distance for contact
SSCT Intensity / filament_ssct_intensity	Float	0.8
SSCT Light Direction / filament_ssct_light_direction	Float 3D	0, -1, 0
SSCT Depth Bias / filament_ssct_depth_bias	Float	0.01	Depth bias in world units (mitigate self-shadowing
SSCT Depth Slope Bias / filament_ssct_depth_slope_bias	Float	0.01	Depth slope bias (mitigate self-shadowing
SSCT Sample Count / filament_ssct_sample_count	Float	4	Tracing sample count
SSCT Ray Count / filament_ssct_ray_count	Float	1	Tracing ray count
Depth of field / group:Depth of field	Category	-
Enabled / filament_dof_enabled	Boolean	Off	Enable or disable depth of field effect
COC Scale / filament_dof_coc_scale	Float	1	Circle of confusion scale factor (amount of blur)
COC Aspect Ratio / filament_dof_coc_aspect	Float	1	width/height aspect ratio of the circle of confusion (simulate anamorphic lenses)
Filter / filament_dof_filter	Choice	Median	filter to use for filling gaps in the kernel - None - Median
Max aperture diameter / filament_dof_max_aperture_diameter	Float	0.01	Maximum aperture diameter in meters (zero to disable rotation)
Max background COC / filament_dof_max_background_coc	Float	0	maximum circle-of-confusion in pixels for the background
Max foreground COC / filament_dof_max_foreground_coc	Float	0	maximum circle-of-confusion in pixels for the foreground
Native resolution / filament_dof_native_resolution	Boolean	On	perform DoF processing at native resolution
Background Ring Count / filament_dof_background_ring_count	Float	0	number of kernel rings for background tiles
Foreground Ring Count / filament_dof_foreground_ring_count	Float	0	number of kernel rings for foreground tiles
Fast-Gather Ring Count / filament_dof_fast_gather_ring_count	Float	0	number of kernel rings for fast tiles
Fog / group:Fog	Category	-
Enabled / filament_fog_enabled	Boolean	Off	enable or disable fog
Color / filament_fog_color	Color	0.5, 0.5, 0.5, 1	fog's color
Density / filament_fog_density	Float	0.1	fog's density at altitude given by 'height'
Distance / filament_fog_distance	Float	0	distance in world units from the camera where the fog starts ( >= 0.0 )
Height / filament_fog_height	Float	0	fog's floor in world units
Height falloff / filament_fog_height_falloff	Float	1	how fast fog dissipates with altitude
Maximum opacity / filament_fog_maximum_opacity	Float	1	fog's maximum opacity between 0 and 1
Fog Color from IBL / filament_fog_color_from_ibl	Boolean	Off	Fog color will be modulated by the IBL color in the view direction.
In-scattering size / filament_fog_in_scattering_size	Float	-1	size of in-scattering (>0 to activate). Good values are >> 1 (e.g. ~10 - 100).
In-scattering Start / filament_fog_in_scattering_start	Float	0	distance in world units from the camera where in-scattering starts
FXAA / filament_fxaa	Boolean	Off
Multisample anti-aliasing / group:Multisample anti-aliasing	Category	-
Enabled / filament_msaa_enabled	Boolean	On
Sample count / filament_msaa_sample_count	Float	16
Custom resolve / filament_msaa_custom_resolve	Boolean	Off
Shadow Type / filament_shadow_type	Choice	PCF	The shadow filtering technique. - PCF: Percentage-Closer Filtered - DPCF - PCSS
Soft shadows / group:Soft shadows	Category	-
Penumbra ratio scale / filament_soft_shadows_penumbra_ratio_scale	Float	1	Globally scales the computed penumbra ratio of all DPCF and PCSS shadows.
Penumbra scale / filament_soft_shadows_penumbra_scale	Float	1	Globally scales the penumbra of all DPCF and PCSS shadows
Screen-Space Reflections / group:Screen-Space Reflections	Category	-
Enabled / filament_ssr_enabled	Boolean	Off	Enable Screen-Space Reflections
Thickness / filament_ssr_thickness	Float	0.1	Ray thickness (world units)
Bias / filament_ssr_bias	Float	0.01	Bias, in world units, to prevent self-intersections
Max Distance / filament_ssr_max_distance	Float	3	Maximum distance, in world units, to raycast
Stride / filament_ssr_stride	Float	2	stride, in texels, for samples along the ray
Temporal Anti-Aliasing / group:Temporal Anti-Aliasing	Category	-
Enabled / filament_taa_enabled	Boolean	Off	Enable TAA
Filter Width / filament_taa_filter_width	Float	1	Reconstruction filter width typically between 0 (sharper, aliased) and 1 (smoother)
Feedback / filament_taa_feedback	Float	0.12	History feedback, between 0 (maximum temporal AA) and 1 (no temporal AA).
Variance shadow maps / group:Variance shadow maps	Category	-
Anisotropy / filament_vsm_anisotropy	Float	0	Number of anisotropic samples to use when sampling a VSM shadow map
Light bleed reduction / filament_vsm_light_bleed_reduction	Float	0.15	VSM light bleeding reduction amount, between 0 and 1
Minimum variance scale / filament_vsm_min_variance_scale	Float	0.5	VSM minimum variance scale, must be positive
Mipmapping / filament_vsm_mipmapping	Boolean	Off	Whether to generate mipmaps for all VSM shadow maps
Adjust To Source Format / adjustToFormat	Boolean	Off	When checked, the copied image is cropped by the Format before being copied. When unchecked, the parts of the copied layer that lie outside the frame are kept. See Adjust To Source Format.