Where did the term Rotoscope come from? A rotoscope was a mechanical device patented by Max Fleischer in 1917. It projected single frames of live action footage onto the animators drawing board. By simply tracing the projected shape, the animator could quickly produce incredibly lifelike drawings. With the passage of time, Rotoscoping or “Roto” for short has become a generic term for drawing shapes to extract, isolate or affect a portion of an image. It is tedious work, but it’s one of the most important parts of the visual effects process.

Go to the Roto tutorials to see how it works.

Image, Silhouette.

When in the Left or Right View of a stereo project, adding a shape creates it in either the Left or Right View.

Go to the Stereo Roto tutorial to see how stereo rotoscoping works.

Go to the Bézier tutorial to see how it works.

To draw a Bézier spline, click on the image to place the first control point and drag the cursor while keeping the mouse pressed to extend the control point’s tangents. Click on the screen to add more control points and when finished adding points, click the first control point that you added to close the shape.

Originally developed by Pierre Bézier in the 1970's for CAD/CAM operations, Bézier splines became the foundation of the entire Adobe drawing model. If you're a user of Adobe products, you've probably used Bézier’s. Bézier splines are defined by control points and tangents. The position of the two tangent handles control the amount of a point's curvature.

Control points are locations on the curve that determine its shape. When a control point is selected, two tangents extend from the control point. These tangents guide the shape of the curve.

Joins can either be corner or cardinal (smooth) joins. When control point tangents are positioned in opposite directions, the join is cardinal. In contrast, corner joins have their tangents positioned at an angle to each other.

Control points have two tangents that can extend from it. The path of the curve through the control point is determined by the length and orientation of the tangents. When you drag one tangent’s handle, the adjacent tangent moves as well. Using keyboard shortcuts while dragging allows you to control how the curve passes through a control point.

Sometimes a control point’s tangents are so close together that trying to modify the tangent results in movement of the control point. If this is the case, you can force the tangents to move without affecting the control point by pressing the Alt key while dragging the tangent. Only the tangent will move, leaving the control point unaffected.

Once you start dragging, the Ctrl/Cmd key can be released.

Go to the B-Spline tutorial to see how it works.

To create a B-Spline, click on the screen to create a control point and add as many points as you like. When you are finished adding points, click the first control point that you added to close the shape.

To create an open shape, add points and press the Esc key or switch to a different tool to finish the shape. Then in the Reshape tool, select the open shape, adjust the Stroke Width to set its thickness and choose either a flat or round Cap Style in the Object Parameters.

In contrast to Béziers, B-Splines don’t use tangents and are created by only using control points. The position of the points, their tension settings as well as their proximity to each other determine the curvature of the shape. B-Splines are somewhat similar to NURBS (non-uniform rational B-Splines) that are used in many 3D modeling packages.

Because the two spline types of have significant strengths and differences, artists are religiously loyal to one or the other. B-Splines create natural curves but are terribly inefficient at defining square corners and require more points to define than a similar Bézier shape. On the other hand, Béziers can easily create corners and use fewer points, but can have an unnatural curved path caused by overly malleable “split” handles. Depending upon your expertise, you may find Bézier splines are more appropriate for square and angular shapes while B-Splines will work better for objects with a lot of curves like a person’s face. Just to upset the status quo, there is a new kid on the block called X-Splines which is explained in the next section. This spline type brings together the best of both B-Splines and Béziers.

In contrast to Béziers, B-Splines don’t use tangents and are created by only using control points and are simple to adjust. The position of the points, their weight settings as well as their proximity to each other determine the curvature of the shape. The weight (amount of pull on a point) decides how sharp or smooth it is. Silhouette employs variable weight intensities on a point by point basis. The weight can be adjusted by hovering over a point and Alt-dragging to the right to create a corner point or Alt-dragging to the left to create a smooth point. Alternatively, you can Alt-click on one of the selected points to cycle through the preset weight settings.

Go to the X-Spline tutorial to see how it works.

To create a X-Spline, click on the screen to create a control point and add as many points as you like. When you are finished adding points, click the first control point that you added to close the shape.

To create an open shape, add points and press the Esc key or switch to a different tool to finish the shape. Then in the Reshape tool, select the open shape, adjust the Stroke Width to set its thickness and choose either a flat or round Cap Style in the Object Parameters.

A X-Spline is an intuitive and easily editable spline format whereby its points can be of three different types: Cardinal, Corner or B-Spline.

The point type can be changed at anytime as well as animated from one type to another. You’ll find that the beauty of X-Splines is that the shape’s points can be a mix of Cardinal, Corner or B-Splines.

X-Splines are created by using control points and then adjusting their weight to coincide with one of the point types: Cardinal, Corner or B-Spline. The point type can even be somewhere in between one point type and another. The weight can be adjusted by hovering over a selected point and Alt-dragging to the right. The weight of the point goes from Cardinal to Corner to B-Spline. Alternatively, you can Alt-click to cycle through the different point types.

You can snap the control points of an X-Spline to edges of an object by using the Edge Snapping tool. Select the points you want to move and press the Alt-S key.

Alternatively, start dragging the points, hold the Alt-S key down and they will snap to the nearest detectable edge. In some cases, the points may not snap to where you want them to. If this happens, adjust the points manually.

Go to the Magnetic Freehand tutorial to see how it works.

Creates a freehand shape with magnetic properties. There are two methods of creating a magnetic shape: Tracing or Strokes.

Add additional foreground and background strokes as needed so that the shape properly surrounds the object.

Press Esc to clear the foreground and background strokes as well as deselect the current shape. You can now create a new magnetic shape or new foreground/background strokes.

Click and drag to create a freehand shape. A freehand shape is exactly like a magnetic shape, minus the magnetic properties. When you are done drawing, click the first control point that you added to close the shape.

Adjusting the Detail reduces points for X-Splines created with the Magnetic Freehand tool, while increasing points for normal X-Splines.

To create a Square, click and drag in the shape of a square and release the mouse.

Go to the Square tutorial to see how it works.

To create a Circle, click and drag in the shape of a circle and release the mouse.

Go to the Circle tutorial to see how it works.

Controls how objects are blended together.

Adds the object to other objects.

Subtracts the object from other objects.

Produces a result where there is a union of pixels from two objects.

Produces a result where a value exists in each object, but not in both.

Looks at the luminance information for each object and selects the value—whichever is brighter—as the result. Darker pixels are replaced while brighter pixels do not change.

Sets the blur level of the selected shape or shapes. The Blur range is from 0-100 and defaults to 0.

Determines the type of blur.

The blur is centered on the edge of the shape, equally blurring inward and outward.

The shape is blurred inward from the edge of the shape.

The shape is blurred outward from the edge of the shape.

Shrinks or grows the shape’s alpha channel. Negative values shrink and positive values grow.

Sets the opacity of the selected shape. The opacity range is from 0-100 and defaults to 100. The interpolation type is set to Hold by default making it easy to turn shapes on and off at particular frames.

Inverts the shape values. The default is off.

Motion blur is the directional blurring of rapidly moving shapes. This parameter turns motion blur on or off for the selected shape or shapes. The default is off. To affect a shape, Motion Blur also needs to be enabled in the Node window. See the Motion Blur section of the Roto Node parameters for more information.

When the Roto node’s Color > Enable parameter is on, this sets the color of the filled shape.

Sets the thickness of an open shape. When used on a closed shape, an outline is created.

Sets the end cap style for open shapes. You can choose from Flat or Round styles.

Determines what channel the shape is rendered into. You can set each shape to Red, Green, Blue, or Alpha. To visualize the results prior to rendering, use View > Channels in conjunction with the Alt-R, G, B and A shortcut keys. The assigned channels can be rendered by connecting the Roto node’s Channels output to an Output node.

Go to the Rendering Shapes to Channels tutorial to see how it works.

Resets all parameters to their default state.

The Position, Scale, Rotation, Corner-Pin and Anchor Point can be adjusted and animated.

Sets the anchor point for the layer transformation. Scaling and rotation operations center around the anchor point.

Sets the horizontal and vertical position of the layer.

Sets the horizontal and vertical scale of the layer. Typing in a negative value will flip or flop the shapes in the layer.

Sets the rotation of the layer.

The layer can be corner pinned by adjusting the Corner-Pin values.

Controls the X and Y position of the Upper Left Point.

Controls the X and Y position of the Upper Right Point.

Controls the X and Y position of the Lower Right Point.

Controls the X and Y position of the Lower Left Point.

The layer’s Surface controls are used for importing or exporting corner-pin trackers, as a reference for an insert/transform, as well as the anchor for the PowerMesh data from Mocha.

Controls the X and Y position of the Upper Left Point.

Controls the X and Y position of the Upper Right Point.

Controls the X and Y position of the Lower Right Point.

Controls the X and Y position of the Lower Left Point.

When tracking data is applied to a layer, the Matrix parameter is used to store the tracker keyframes.

Stereo Offset, which only shows up when using stereo sources, is used to align the offset between Left and Right Views of stereo clips and is set to animate by default. This value is automatically set when using the Viewer’s Stereo Align mode’s on-screen control.

Controls how layers are blended together. See the Blend Mode section for more information.

Sets the blur level of all shapes within the layer. The Blur range is from 0-100 and defaults to 0.

Determines the type of blur.

All shapes in the layer are blurred equally inward and outward centered on the edge of the shapes.

All shapes in the layer are blurred inward from the edge of the shapes.

All shapes in the layer are blurred outward from the edge of the shapes.

Inverts all shapes within the layer. The default is off.

Determines which channel the shapes in the layer are rendered into. You can set a layer to Red, Green, Blue, or Alpha. To visualize the results prior to rendering, use View > Channels in conjunction with the Alt-R, G, B and A shortcut keys. When ready to render into the assigned channels, use the Actions > Render Shapes to Channels action.

Go to the Rendering Shapes to Channels tutorial to see how it works.

Resets all parameters to their default state.

Click the Search icon, enter an object name into the search field and it will appear.

For instance, to show all right side objects, just type /r. To show all left side objects starting with the word deer, type deer/l. Entering deer/rb would match all right side B objects that have the word deer in the name.

Disable the Search icon when done to return to the normal Object List window view.

You can rename an object by either:

Clicking the name of an object selects it. Multiple objects can be selected by Shift-clicking on them. Ctrl/Cmd-clicking will add to the selection of objects that are located at different levels of the Object List while Ctrl/Cmd-clicking an object after it is selected will deselect it. In addition, if you Shift-click a color pot in the Object List, you select all objects of the same type and color.

Objects can be moved by dragging and dropping them to a new position in the list or within another layer.

If expanding or collapsing a selected layer using +/- icon, then any other selected layers with the same parent will also expand or collapse. This can be combined with Shift to expand or collapse children as well.

Clicking on the Opacity icon toggles the shape off (0% opacity) or on (100% opacity) on the current frame. Keyframing is also enabled and the animation type is set to Hold. This allows you to conveniently turn the shape on and off.

Shift-click on the color pot to select all objects with the same color.

Selects the font type.

Sets the font size.

Enables bolding.

Enables italics.

Enables underlining.

Sets the text color.

Clears all text.

The Layer’s View, Blend Mode, Invert, Color, Lock, Visibility and Notes icons have similar functionality to the ones used by other objects.

Once you draw a shape, there are various ways to modify it. As the object changes over time, you will need to adjust the shape so that it exactly matches the object. While the Transform tool affects an entire shape or a group of shapes, you can use Reshape to modify individual points. See Reshape for more information.

Go to the Transform tutorial to see how it works.

If there are any selected layers while in the Transform tool, pressing Q, W or E and then clicking and dragging will translate, rotate or scale the selected layers. If there are no selected layers, you must select a shape or group of shapes before clicking and dragging. Also, the position of the cursor when W or E is pressed sets the anchor point for the rotation or scale operation.

Point Transform allows you to position, move, scale, rotate, shear or corner-pin a selection of points using the on-screen controls.

Instead of a bounding box with on-screen controls around the shape, the bounding box is around the selected points.

Selected shapes are moved using one of three methods:

In either case, you can constrain movement along an axis by pressing the Shift key while dragging the outline of a shape. The movement is constrained by the first direction (horizontal or vertical) that you drag in.

To proportionately scale, hold down the Shift key when resizing.

Now, when you rotate or scale the shape, notice how the shape rotates and scales around the Anchor Point.

Shapes can be cut(Ctrl/Cmd-X), copied(Ctrl/Cmd-C) and pasted(Ctrl/Cmd-V).

A shape or selection of shapes can be easily deleted using none other than the Delete key.

You can also use the arrow keys to nudge the layer. See Nudging for more information.

Reshape modifies shapes by adjusting their control points and tangents.

When a control point is selected, it’s control points and tangents become visible and can then be fine tuned.

Go to the Reshape tutorial to see how it works.

While in Magnet mode with multiple selected control points, you can Shift-Alt click/drag in the Viewer, not on any control points or shapes, and that becomes the magnet pull position instead of using a point.

Adding control points assists in tracing more detailed images. To add a control point, Alt-click on the selected shape. Don’t click on an existing control point because it will move it instead.

You can delete control points in one of two ways. Select the control point and press the Delete key or right-click on the control point and select Delete from the Reshape pop-up menu.

Control points are easily moved. Just select one or more control points and drag one of the selected points to a new location. You can also use the arrow keys to nudge the points. See Nudging for more information.

All points or a selection of points can be numbered. To number one point or a selection of points, right-click on a selected control point and select Tag. To see all points numbered, go to File > Preferences > Shape on Windows and Linux or Silhouette > Preferences > Shape on Mac and set the Number Points preference to All.

Click on a single point to select it.

Ctrl/Cmd-click a point to toggle the selection.

Select a range of points by Shift-clicking start and end points.

Extend the current selection by Shift-clicking an unselected point. The selection is extended by finding the nearest point of the selection.

Click and drag a rectangle over the control points to be selected. This is the default behavior.

Control points can be selected by drawing a freehand polygon. Ctrl/Cmd-drag in the Viewer to begin the selection. Once the selection is started, release Ctrl/Cmd and complete the selection.

A shape’s points can be feathered to create variable edged blurs on a point by point basis.

To drag shape and feather points simultaneously, press Shift while dragging the shape point.

The display of the feather handles can be hidden using Alt-F.

Multiple selected points can be dragged simultaneously and by a proportionate amount. To move the shape and feather points simultaneously, press Shift while moving the shape point.

Go to the Feather tutorial to see how it works.

To modify multiple selected points simultaneously, do not right-click on one of the selected points. Right-click on an open area of the Viewer instead.

Inverts control point selection state.

Deletes the selected control point.

Breaks the shape at the selected control point.

Closes an open shape.

Resets the attributes of the selected control point.

Both tangents are adjusted to one quarter the distance and in the direction of their adjoining control points.

Creates a corner point.

Creates a smooth point where the curve passes smoothly through it.

Creates a corner point.

Creates a smooth point that is determined by the surrounding points.

Creates a corner point.

Creates a smooth point where the curve passes smoothly through it.

Creates a smooth point that is determined by the surrounding points.

Centers the control point between its adjoining control points.

Tags control points for point numbering. For point numbers to be displayed, the Shape > Number Points preference must be set to Tagged.

Groups a selection of points for fast selection when editing in Normal/Magnetic Reshape modes as well as Point Transform mode. Each group is assigned a color as defined in the Colors > Point Groups preference.

With points selected, choose group 1-8. If points are already in a group, a check mark appears next to the number. Also, the group number is underlined if a group already exists.

Removes selected points from all groups.

Remove all grouping for the selected shape.

Takes the point and “re-tweens” it using the surrounding keyframes, and puts them where they would be if you hadn't moved them on that frame. Useful for simulating an “unkey” of selected points (the shape remains keyed of course).

Reverses the order of all control points on all keyframes. Useful for warping and morphing, Reverse is used when you are trying to join two shapes where the control points go clockwise in one and counter-clockwise in the other. In this case, the correspondence points would iterate in different directions and the lines would cross each other. Reversing the points of one of the shapes would make them go in the same direction and the correspondence would be correct.

Selects all open shape start or end points.

Similar to Split Edit in non-linear editing systems where a clip is split in two, Split Shapes is useful for shapes that transform from simple to complex and vice-versa.

The original shape is duplicated and named Shape-copy. If there is an active selection, the new shape is created based on the selected points. At the current frame, the original shape is set to 0% opacity and the copied shape set to 100%. Both shapes are placed in a layer named Shape-Compound Shape. In editing terms, you are cutting from one shape to another.

Extracts the selected points of a shape into a new closed shape.

Copies selected points from a source shape between the selected points of a target shape.

Collapses a selection of points into a very small area which can later be distributed between surrounding points. This is useful for shapes whose complexity changes over time.

Drag-selecting a box around the collapsed points selects them.

Evenly distributes selected points.

In the Reshape tool, closed shapes can be opened using the Break option from the control point pop-up menu. To break a shape, select a point and right-click on it to open the pop-up menu. Choose Break and the shape opens at the selected control point.

Shapes can be closed or extended with just a few mouse clicks. It is necessary to use the Reshape tool to perform the following actions.

Select one of the control points at the end of the shape and Alt-click on the other end point to close the shape.

Select a control point at either end of the shape and Alt-click somewhere off the shape.

In the Reshape tool, open shapes can be joined together. To join open shapes, select one of the end points of the source shape by clicking on it. Next, make sure both shapes are selected in the Object window. Finally, Alt-click on one of the end points of the target shape. The two shapes are now joined together as one.

You may convert your B-Splines or X-Splines to Bézier splines at any time. To convert a B-Spline to a Bézier spline, select a B-Spline or X-Spline and choose the Edit > Spline > Convert to Bézier menu item.

Once you select Convert to Bézier, a couple of things happen. The visibility of the B-Spline or X-Spline is toggled to off. A new Bézier spline is created and is named the same as the B-Spline or X-Spline but has an * at the end of the name.

Shapes, control points and layers can be nudged using the Arrow keys. One press of the Arrow key moves 1 pixel. Using the Shift key in conjunction with the Arrow keys moves 10 pixels. Ctrl plus the Arrow keys moves one tenth of a pixel. These nudge values can be changed in the Preferences menu located in File > Preferences > Nudging on Windows and Linux or Silhouette > Preferences > Nudging on Mac. Holding the Arrow keys down moves the selection continuously.

Go to the IK tutorial to see how it works.

Arm, Elbow and Wrist are layers while Upper Arm, Forearm and Hand are shapes located in the appropriate layer.

Click-drag a bone or shape to rotate the shape around the parent’s anchor point. Click-dragging a joint uses IK to rotate all joints up the chain. Alt-click-drag a joint to move the joint while attempting to keep other joints in place.

When animated, keyframes are automatically set for each layer’s rotation parameter. However, you can explicitly set the layer rotation keyframes by right-clicking on a selected bone and choosing Add Key.

Go to the Tracker node to see how it works.

The keyframe adjustment can also be weighted based on a specified Fade In or Fade Out time period. Keyframes within the fade areas are adjusted less than those that are not. The MultiFrame Start and End as well as the Fade In and Fade Out parameters can be set in either the Timebar or Timeline.

Sets the start of the MultiFrame range.

Controls the strength of the MultiFrame adjustment. For instance, if an object is slowly drifting, the strength of the adjustment can be faded in or out.

Sets the first frame where the MultiFrame adjustment is 100%.

Sets the last frame where the MultiFrame adjustment is 100%.

Sets the end of the MultiFrame range.

Go to the MultiFrame tutorial to see how it works.

Add Weighted Key creates a keyframe at the current position which changes the timing of the shape animation between two surrounding keyframes. Park between two keyframes, select a shape, press Alt-K, and drag the slider to adjust the weight.

Hold Alt over a motion path marker and the frame number is displayed in the status bar. Click on the marker and the Timebar will jump to that frame.

Hold Alt over an onion skin shape and the frame number is displayed in the status bar. Click on the shape and the Timebar will jump to that frame.

Determines the amount of frames used in the onion skin.

The Onion Skin Frame Range Backward numeric field (the first one) displays the frames before the current frame.

The Onion Skin Frame Range Forward numeric field (the second one) displays the frames after the current frame.

As an alternative to using the pre-defined colors for the Motion Path, Motion Blur and Onion Skin, you can use the default shape colors.

Silhouette, Mocha Pro, and Shake 4.x SSF shapes can all be imported into Silhouette using the File > Import > Shapes menu.

Go to the Import tutorial to see how it works.

Silhouette, Fusion, gMask (Flint, Flame, Inferno), Nuke and Shake 4.x SSF shapes can all be exported using the File > Export > Shapes menu.

Go to the Export tutorial to see how it works.

The Silhouette Shape Import/Export Plug-in for After Effects is required to convert Silhouette Shapes into After Effects masks. When importing Silhouette Shapes, the following shape parameters transfer into After Effects: opacity, blur, shape color, shape name, invert and locked states, transfer mode and per-shape motion blur state.

Go to the After Effects section of the Import tutorial to see how it works.

The Silhouette Shape Import/Export Plug-in for After Effects is required to convert After Effects masks to Silhouette Shapes. When exporting After Effects masks, the following shape parameters transfer into Silhouette: opacity, blur, shape color, shape name, invert and locked states, transfer mode and per-shape motion blur state.

Go to the After Effects section of the Export tutorial to see how it works.

There are three Nuke shape export options--Nuke 5 Shapes, Nuke 6.2+ Shapes and Nuke 9+ Shapes.

Nuke 5 Shapes (for Nuke 5.1v3 and above), Silhouette shapes and their blend modes are exported. Silhouette will also export Hold and Linear key frames for Opacity. When exported, tracker and shape data are baked together.

Nuke 6.2+ Shapes will export the shape opacity, blend mode, invert, motion blur, outline color, shape blur (inner and outer blur only) as well as the layer and node blur and motion blur.

Nuke 9+ Shapes will export open/closed shapes, the shape feather, opacity, blend mode, invert, motion blur, outline color, fill color, shape blur (inner and outer blur only), and visibility. The layer’s blur, visibility and motion blur are also exported in addition to the node blur. You can optionally fully bake shape and transform data, convert opacity to Nuke Lifetime, and export the Silhouette project name as determined by the Nuke 9+ Shapes preferences.

Silhouette will export the shape name, locked/unlocked state, opacity, visibility, motion blur on/off, shutter angle, and shutter phase. In addition, subtractive shapes import into Shake as black in the RGB channels with a white alpha channel.

When the Roto node is being edited in the Trees window, parameters specific to the Roto node, such as Motion Blur, can be adjusted in the Node parameters.

Inverts the alpha channel.

Blurs the alpha Channel. The blur range is from 0-100 and defaults to 0.

Renders shapes as outlines or filled with color in the RGB output. The color is set using the Shape > Color parameter.

Activates the color parameters.

Sets the thickness of the shape outline.

Sets the opacity of the color fill. You can also change the opacity of each shape separately using the Shape > Opacity control.

Motion blur is the directional blurring of rapidly moving shapes. Enables Motion Blur for the Roto node.

Go to the Motion Blur tutorial to see how it works.

Turns Motion Blur on or off. The default is off. Leave this turned off for faster interaction while editing shapes.

Determines how long the camera shutter stays open when a picture is taken-- higher values create more motion blur. The range of the Shutter Angle is 0-720 and defaults to 180. Measured in degrees, it simulates the exposure of a rotating camera shutter. The shutter angle uses the footage frame rate to determine the simulated exposure. For example, a shutter angle of 180 degrees (50% of 360 degrees) for 24fps footage creates an effective exposure of 1/48 of a second. Typing 1 degree applies almost no motion blur, and typing 720 degrees applies a high degree of motion blur.

Offsets the point in time, either forward or reverse, when the shutter opens. The range of the Shutter Phase is -360 to 360 and defaults to -90.

Renders intermediate frames equal to the Motion Samples value and accumulates them, one over the other, on a single frame. The higher the number, the smoother the motion. The Motion Samples range is from 1-256 and defaults to 16.

Controls whether roto shapes are antialiased on their edges.

Outputs the foreground input and generated matte.

Outputs a composite of the foreground over a solid color based on the Roto > Composite Color preference. The default color is gray.

Outputs a composite of the foreground over the background based on the generated matte.

Renders shapes to different channels of an RGBA image as defined by the Shape > Channel parameter.

Outputs tracked layers and shapes for use in nodes with Data inputs.