4
Video Editing

Once a video signal has been digitised, it can be considered as simply a time-based series of still digital video images. From within video editing applications, single or multiple effects can be applied to all the frames within a movie or just to a selected clip within the movie. For more detailed editing, a selected clip can be exported to an image editing application as a ‘frame stack’ – a format which separates the clip into its individual frames. After editing of the individual frames is complete, the frame stack can be converted back to video clip format for reimporting back into the video editing application.

By temporarily exporting a frame from its location within the timebase to a digital photoediting or painting application like Photoshop or Painter, the whole array of techniques available within such applications can now be brought to bear on the contents of the frame, i.e. video editing is now possible not only at the individual frame level, but even at an individual pixel level within a frame!

It is through this level of control that digital video editing has made a quantum leap beyond the capabilities of analogue film or video editing. Now the pixels comprising individual video frames can be manipulated in much the same way that still images are manipulated in image-editing programs.

The reverse process is also possible, i.e. still images created in any drawing, photoediting, painting or 3D application – including those employing sophisticated atmospheric effects, procedural textures, etc. – can be inserted into, or overlayed on top of, a video stream. It is this ease of mixing and matching still images with video and sound which provides the basis for a breathtaking range of digital multimedia production possibilities.

In Chapter 3, we already saw a few examples of the software available for adding simple transitions, titles and motion. In this chapter we shall explore in more depth the remarkable range of video and audio special effects which digital video brings to the desktop.

Editing in the Host Applications

The editing techniques provided within applications like Premiere, Lumiere or MediaStudio fall basically into five categories – Transitions, Text effects, Motion effects, Video overlay and Filters. The following sections cover each of these in turn.

Transitions

The most common transition between clips is a cut – an instantaneous switch from one clip to the next one. The term derives from film editing, where edits are carried out in a ‘cutting room’ by literally cutting required sequences from reels of film and then splicing them together to make up a new reel.

In digital editing in the construction window, cutting between clips in a videoediting program is achieved by simply arranging the clips head to tail on the same track in the construction window (if there is no transition effect and clips overlap, then clips in higher numbered video tracks play over the top of clips in lower numbered tracks).

Used correctly, transitions can provide an elegant and visually pleasing bridge between clips, adding value to the finished composition. Used incorrectly, they can interrupt the flow of visual information and distract attention from the project content. Because so many different transitions are available, the temptation is to use a variety, but for most projects, less is better.

In the majority of cases the simple cut from clip to clip works perfectly well and no transition is required; continuity is provided by the sound clip on the audio track. When a smooth and unobtrusive change from clip to clip is required then a simple Crossfade – which causes the first image to fade out as the second image fades in – works well (Figure 4.1). The choice of transition depends on the subject mood of the images on the clips being joined and on their contents; a fancy Checker-Wipe would not be appropriate between two slow-moving romantic scenes, but might work well in linking two motor-racing clips. Some transitions simulate physical events, such as Side-Slide which visually suggests the sliding back of a door, or Center-Slide, which suggests the opening of curtains. Counter-Clock simulates the movement of a hand around a clock, while Random-F/X causes the top image to break up and disappear in random fashion, revealing the underlying image. Transitions like these can help convey a particular message; for example, Random-F/X can suggest a flashback in time.

Figure 3.11 showed thumbnails of the transition effects provided with MediaStudio. These were divided into a number of categories with names like Push, Slide, Wipe, Stretch, Peel, Clock, 3D, Roll and Build, which help to suggest the nature of the transitions within each category. Media Studio’s Transition Options dialog box (Figure 3.12) provided the means of adding borders and soft edges to the clips during the transition process.

Lumiere provides a wide range of transition types, listed in alphabetical order (Figure 4.2). Thumbnails demonstrate the effect of each transition and a descriptive message appears in the text box in the bottom-left hand corner when a transition type is selected. Figure 4.3 shows Lumiere’s Transitions dialog box.

By default, the second track which the transition applies to is not visible (0%) at the start of the transition and is fully visible (100%) at the end. Clicking Swap Sources changes the order in which the transition is applied to the two clips and switches the clips shown in the Start and Finish preview boxes. For transitions with a direction of movement, the direction settings can be changed using the buttons in the Direction section of the dialog box. The Border slider adjusts border thickness and a colour for the border can be chosen by clicking on the colour swatch. Figures 4.4 and 4.5 show a few examples.

As well as straightforward application between two sequential videoclips, a transition clip can be used to create other effects – for example, to create ‘video within video’. First a video transition clip is placed along the entire length of two clips on tracks Va and Vb (Figure 4.6a), then, in the Transition Options dialog box, the Start and End transition sliders are set to the same value – 50% in the example shown (Figure 4.6b). The result is that the clip of the spacecraft plays in a window, while the clip of the planet plays in the background. The percentage figure set on the sliders determines the size of the window, a smaller percentage producing a smaller window.

Although there are several different methods which can be used to create a split screen, using a transition between clips on two video tracks is probably the easiest. In Lumiere (Figure 4.7a), video clips of a fire and a blizzard were first placed on video tracks A and B and a then a Wedge Wipe transition was dragged from the Transitions Roll-Up and placed on the transition track between the two clips (Any Wipe transition can be used for this effect). After double-clicking on the transition to display the Wipe Settings dialog box, the sliders below the Start and Finish frames were dragged until they were each set at 50%, splitting the screen in half, so that, when the video was run, the two scenes played back side by side in their separate halves of the screen.

Text effects

Although the most common use of text in a video project is to ‘top and tail’ the moving video with titles and credits, text also plays a wider role in training and promotional videos, where bullet text charts, tables or annotated flowcharts, for example, may be combined with video and sound to create a multimedia format.

For whatever purpose the text is being used, the first consideration should be clarity and legibility; overlaying blue type on a blue sky will not get the message across; neither will the use of a decorative script typeface at a point size which is illegible at normal viewing distance for the planned viewing device. Over capitalisation of type should be avoided; lower case type is easier to read. Brevity is also a virtue – a short, sharp, two-or three-word title is preferable to a rambling description which the viewer will probably not read to the end anyway.

The style of typeface should also be chosen to complement the video material to which the text applies. The same face will not be suitable for a business video, a pop video and a lyrical countryside travelogue. Thankfully the days of simply choosing between Times Roman and Helvetica are long gone; the titling utilities provided with video applications can access any typeface available on, or accessible by, the user’s system, including hundreds, if not thousands of typefaces bundled with graphic applications like CorelDRAW or Illustrator. The range is truly spectacular and therein lies another potential pitfall – the temptation to use a variety of different faces within the same project. Generally, the fewer the typefaces used the more cohesive and integrated the result.

Traditionally, typefaces with similar attributes have been divided into five categories – Serif, Sans Serif, Script, Decorative and Machine. Figure 4.8 shows a few examples in each category. Selection of the ‘right’ typeface for a particular project is highly subjective; in general, Serif and Sans Serif faces provide a safe default choice, but other faces can give a project extra impact, for example by choosing a typeface with a recognised visual association, like those shown in Figure 4.9.

The typeface is just one of the text variables which can be set from within videoediting applications. Other user definable parameters are listed below:

  Text alignment – left, centred, right or justified

  Typeface size, style (normal, bold, italic or bold-italic) and type colour

  Outline colour

  Shadow offset and shadow colour

  Text transparency (uniform or gradient)

  Background (matte or an imported image or a video clip freeze frame, which will show behind the text as a still image)

Using a suitable combination of the above parameters, a text clip can be prepared and placed on a video track, where it plays like a video clip. Placing it on a superimposition track allows the background to show behind the title. The transparency of superimposed text can also be increased to let the underlying video show through the text itself.

While offering the impressive range of text editing options described above, videoediting applications cannot compete with the text manipulation features of drawing or painting applications like CorelDRAW, Freehand or Photoshop. When a project requires something special, one of these applications can be used to explore an even wider range of options. A page is simply set up in the application to match the form factor of the video clip; the required effect is created and the page is then saved in bitmap format ready for importing to a video track or superimposition track. Additional variables which can be manipulated from within drawing applications include:

  Fancy text fills (gradients, textures or patterns)

  Text rotation to any angle, skewing or mirroring

  Extrusion and perspective

  Embossing

  Fitting text to a path or a shape

  Blending

  Fitting text to curve

  Node editing of text characters

  Masking

  Combining text with graphics to create ‘word pictures’

These variables can be applied individually or in combination to create a dazzling range of effects. A few examples, taken from an earlier book – Digital Graphic Design – are shown in Figure 4.10.

.

A technique which can be used to provide information about what is appearing on a video track in an unobtrusive way is to fade the text into view and then to fade it out after allowing time for it to be read by the viewer. Figure 4.11 shows how this fading effect is applied in Lumiere.

After creating a text clip using the titler and saving it to the Media Catalog, the clip is dragged on to a superimpose track, in this case S7. Clicking anywhere along the lower border of the clip opens the fade window immediately below the clip. By default, fade is off, as represented by a solid line running along the top of the fade window; this indicates that the superimpose clip is 100% visible. Three tool buttons to the left of the window are used to edit the fade curve. The Fade Control tool can be used to move the entire fade control line up or down. The Add/Delete tool is used to add or remove control points anywhere along the line. The Adjust Control Point tool is used to manipulate the positioning of control points once they have been created. In the example shown in Figure 4.11, the text fades from 0% visibility to 80% visibility, where it remains for about 2 seconds before it fades back to 0%.

In Chapter 3 we saw how scrolling titles could be created using the Pan feature in Lumiere. Simple text animation is also possible from within Studio 200’s titling facility (Figure 4.12). Clicking the Animation button opens the dialog shown in Figure 4.12; the Type of Effect drop down list in the Entrance group box offers a number of ways in which text (and any associated graphics) can be animated on to and off of the screen:

  Wipe starts to reveal text starting from one edge and wiping across the text until it is all visible

  Slide causes text to enter the screen from one side or corner and then to slide to or from the centre of the video frame

Repeated clicking on the Entrance or Exit arrow buttons rotates through a range of directions which can be applied to the animation effect and sliders are provided to adjust the duration of the effect. Fade in and Fade Out buttons are also provided. Clicking on View Entrance and then View Exit shows a preview of the animation on an attached TV monitor.

In Chapter 5 we shall look at more sophisticated ways of animating text as well as other objects.

Motion

A general rule for the use of motion effects is the same as that for the use of transitions – they should be used sparingly to avoid inflicting motion sickness on the viewer! When a motion effect is used, it should ideally complement the content of the clip to which it is being applied – e.g. the use of a 3D Sphere path to end a clip about a sailing ship circumnavigating the globe, or the use of a Zoom Out effect to mark the end of an episode.

The basic motion features provided within videoediting applications were described in Chapter 3 and can be summarised as follows:

  Moving clips along two-dimensional paths

  Moving clips around imaginary three-dimensional surfaces

  Rotating clips in two dimensions

  Zooming in or zooming out

  Distorting shape, e.g. by applying perspective

  Accelerating or decelerating along a path

  Compressing or expanding the duration of a clip to create slow or fast motion effects

  Reversing the direction of a clip

  Skipping frames to create jerky, staccato movement

  Panning to create the impression of motion in a still image clip

  Applying presets based on a combination of the above variables

Figure 3.38 showed Premiere’s dialog box which is used to apply motion effects such as rotation, zoom and distortion to video clips. Figures 4.13 and 4.14 show some examples of the results which can be created.

Lumiere also includes a menu of motion presets, including:

  Bounce In – clip drops into view and appears to bounce before settling on the bottom of the frame

  Falling – clips falls from frame out of sight

  Flip – clip tumbles and flips over from left to right

  In and Out – clip increases and decreases in size as it traverses across the frame

  Loose – clip wobbles as if coming loose from its fixing and then drops out of sight

  Ricochet – clip moves around the frame as if bouncing off the sides

  Slidehold – clip slides into view, pauses, and then slides out of view like a slide displayed on a projector

  Sweepin – clip follows a curved path, starting from a small size in a corner of the frame and zooming along the path until it is centred, at full size

  Swing – clip appears to swing to and fro’, as if hinged along its top edge

  TV – clips shrinks to a horizontal line and then a dot as if viewed on a television monitor being switched off

  Walking – clip is distorted to look like a playing card walking across the frame from left to right

Video Overlay Effects

A technique used to powerful effect both on television and in the cinema, the overlay employs colour keying to make part of a video clip transparent, revealing an underlying clip in the transparent area. Because overlaying a clip basically conceals or reveals parts of the underlying clip, the clip used as an overlay is often referred to as a mask. The most common use of this technique is in television news or weather presentation; if the overlay clip on a superimpose track showing the presenter is shot against a solid colour background – usually blue – and the blue is selected and made transparent, then whatever image appears on the underlying clip on a video track will be revealed, so that the presenter appears to be sitting or standing in front of this virtual background image. Figure 4.15 shows an example. In MediaStudio, a shot of a great white shark set against a blue background (a) was placed on superimpose track VI and a shot of an underwater scene (b) was placed on video track Va, as shown in (c). After clicking on the shark clip, Overlay Options was selected from MediaStudio’s Clip dropdown menu, opening the dialog box shown in (d). Selecting Color Key from the Type menu displayed an Eyedropper tool which was used to click on the blue background in the Overlay Clip window. The effect of this was to render the background transparent – i.e. the blue is said to be ‘keyed out’ – revealing the clip of the underwater scene on track Va and merging the shark into the underwater scene.

Media Studio provides a variety of options for keying in the colour of an image with controls offered by the Type and Factor boxes as well as the Similarity group box. The best option to use depends on the type of clip selected and the effect being sought. As a rule of thumb, keying on the colour of an image works well with solid colour areas such as blue screens. If the overlay clip is greyscale, then keying is done on the grey. The luma key is useful for keying on the brightness levels of an image, while the chroma is best for keying on the colour values of an image. If the overlay clip is an image, its alpha channel can be used for keying if it has one.

The overlay clip used as a mask can, itself, be modified by using image or video mattes which work also very much like masks, except that they are used t o cover or reveal certain parts of the overlay clip itself, instead of the clip in the timeline, effectively ‘masking the mask’. A typical example of use of a video matte seen on TV shows like the BBC's popular Top ofthe Pops is the silhouette of a dancer appearing over a background, and the silhouette of her body containing another image or video.

The MediaStudio Overlay Options dialog box offers a wide range of possibilities. By simply adjusting points along the Transparency curve for example (Figure 4.16a), the overlay clip can be made to change progressively from being fully visible, to being 50% transparent (Figure 4.16b), to being invisible, or vice-versa.

Other variables set from the dialog box are summarised below:

  The Type menu provides options for specifying the basis for selecting the overlay mask

–  Color Key works well when working with solid colours in the image, such as blue screens

–  Chroma Key works with the colour values in an image and is used to specify a colour, or range of colours. The eyedropper tool can be used to select a colour from the clip, or a colour can be selected by clicking on the colour swatch to open the colour palette. The Similarity slider is used to select a range of related colours. When the Similarity slider is set to zero, only the selected colour is made transparent. As the slider is moved to the right, the tolerance increases so that similar colours are made transparent

–  Luma Key works on the brightness values in an image and is useful for making dark or light areas of an image transparent

–  Alpha channel can be used if the overlay clip is an image and has been saved, e.g. from Photoshop, with an alpha channel; keying takes place on the alpha channel. An example is shown in Figure 4.17. The alpha channel appears in the Overlay Clip window and the result is that the area outside the channel is masked, allowing the underlying clip to show through

  Mask allows the use of a video or image file as the source for the overlay. In Figure 4.18, the overlay clip (the face) is masking an underlying clip of our friend, the motorcyclist. By selecting Mask/Overlay Clip and then using the eyedropper to click on the grey in the face, we can actually ‘mask the mask’ i.e. the underlying clip plays through just the grey areas of the face. Quite spectacular results can be achieved by using this effect. As well as using colour to moderate the effect of the overlay clip, by clicking on Mask/Image Matte, an image can be imported – such as the checkerboard pattern in Figure 4.19 – and used to edit the overlay mask. Even a video clip can be used to edit an overlay mask as shown in the example in Figure 4.20. The video clip to be used as a mask is imported by clicking on Mask/Video Matte and importing the video file into the Overlay Clip window; previewing the effect shows that the video dynamically edits the transparency of the overlay clip as the movie plays

  The Similarity sliders specify the range of key colours relative to the selected one for the overlay. The higher the percentage, the more colours are used as the overlay. Two slider bars correspond to the starting and ending frames for the clip

  Factor (RGB colour, Luma or Chroma) determines which colour components of the underlying clip are dominant

  Soft Edge (None, Small, Medium or Large) controls how strong the transition is between the overlay and the clip. Large results in a less pronounced transition

  Invert Overlay Area makes all colours except the selected one key colours

Alpha Channel mapping can be applied to situations requiring highlighting of a feature within a video clip, such as a figure appearing in a security video. Figure 4.21 shows an example. A still image clip is first created, in an application like Photoshop or Illustrator, containing only an alpha channel – in this case a simple circle – and placed on track V1, with the clip containing the feature to be highlighted placed on track Va. By setting the Type to Alpha Channel and adjusting the transparency to 50%, the head and shoulders of the figure in the Va clip becomes highlighted in the area defined by the alpha channel. By applying a simple 2D Path to the overlay clip (Figure 4.22), the highlighted area can be made to track the figure as it moves within the video clip.

Via its Transparency dialog box (Figure 4.23), Lumiere offers seven different key types – Chroma, Luminance, RGB difference, Blue screen, Green screen, Image matte and Difference matte, providing an even wider range of possibilities than MediaStudio, when the key types are used in combination with the twenty different modes selectable from the Merge mode menu in the dialog box. Lumiere’s Merge modes are similar to those found in image editing applications like Photoshop; they determine the way in which the hue, saturation and brightness values of two overlayed images interact with each other.

The Blue screen key type makes the colour chroma blue transparent (chroma blue is a solid blue containing little or no red or green, corresponding approximately to PANTONE 2735, and is effective in isolating skin tones). Filming a person or group of people against a chroma blue background is a common technique used to display the person or group against a virtual background.

The Green screen key type is the same as Blue screen, except that it makes chroma green transparent (chroma green is a solid green containing little or no red or blue, corresponding approximately to PANTONE 354, and is used to isolate reds and blues from a background).

The Difference matte key type makes the identical areas of two clips transparent and shows only the areas which are different. When Difference matte is selected as the key type in the Transparency dialog box, a clip must be selected via the Browse button to be used as the matte for the superimposed clip.

Premiere offers the most impressive number of different key types, available from the dialog box shown in Figure 4.24. As well as the types described already are White Alpha Matte and Black Alpha Matte. White Alpha Matte is used to superimpose an image which contains an alpha channel and which has been created on a white background (it eliminates the halo-like remnants of white around the edges of the foreground image and is useful for superimposing titles which have been created on a white background). Black Alpha Matte is used to superimpose an image which contains an alpha channel and which has been created on a black background. The Multiply key type has the effect of casting the superimposed clip on to the bright areas of the underlying image. The Track Matte uses a video clip to create a moving mask.

  The Similarity slider is used to select a range of colours to be made transparent. To select a range of colours similar to the one in the colour swatch, the Similarity slider is dragged between None and High; the higher the setting, the broader the range of colours selected

  The Blend slider smooths sharp transitions in colour by creating a gradual change in opacity in the pixels between the two colours

  The Threshold slider is used to adjust the amount of shadow in a superimposed clip

  The Cutoff slider allows adjustment of the shadow detail with the luminance and chroma keys

  The Reverse Key option is used to reverse the transparent area – for example, from the area inside a matte to the area outside a matte

  The Drop Shadow option applies a 50% grey shadow slightly below and to the right of the transparent portion of the clip

  The Mask Only option creates a black-and-white or greyscale mask from the transparent portion of the clip – useful when exporting a clip to Photoshop for retouching

  The Smoothing option creates soft edges where colour transitions occur throughout the superimposed clip. Options are None, Low, or High

The three icons under the Sample window in the Premiere’s dialog box (Figure 4.22) provide different methods of previewing transparency effects. Using the first icon, black or white can be chosen to fill the background. The second icon places a checker-board pattern in the background and the third icon allows the actual background clip to show through. The handles at the four corners of the Sample window can be dragged to reduce the visible area of the clip on the superimpose track. This can be used as another method to create a split screen (Figure 4.25).

Video Filters

The use of filters, special effects which alter the appearance of still bitmapped images in photoediting or painting applications like Photoshop or Painter, has widely extended their artistic repertoire. Although as yet less sophisticated in their range, filters based on the same technology have been ported to videoediting applications.

In Media Studio, the intensity of a filter can be varied at the start, end, and intermediate stages of a clip. In theory, each clip in a project may have up to twenty filters applied at one time.

Figure 4.26 shows Media Studio’s Video Filters dialog box which lists available filter types in the left-hand window. After selecting a video clip, clicking on Clip/Video Filters opens the dialog and the required filter – for example Emboss – is selected using the Add button. Clicking the Options button opens the window shown in Figure 4.27a, where the effect can be previewed and adjusted. The Range button opens window (b) where handles can be dragged to define the area of the clip affected by the filter (c).

Figure 4.28 shows the full set of filters supplied with Media Studio. The effects of applying a number of these to the same video clip are shown in Figure 4.29:

  Average: Gives the clip a softer look by moving the colour values of all pixels towards those of the average pixel

  Find Edges: Displays the clip with outlines around the edges of objects defined by strong differences in pixel colour values

  Fish Eye: Gives the clip a wide angle camera lens effect

  Invert: Converts the clip to its complementary colours, creating a negative effect

  Mosaic: Breaks each frame into blocks closely matching the average values of all pixels in these blocks

  Punch: Displays the clip as if its centre had been pushed out

Media Studio’s filters, like its transitions, can also be dragged from a window (Figure 4.30) displaying thumbnails of each effect on to the video clip to be filtered. This automatically opens the preview window in Figure 4.27a for editing if required.

Corel’s Lumiere offers an impressive list of 58 different filter types. Like Media Studio, Lumiere provides a Video Filters dialog box (Figure 4.31) where filters can be selected and their effects previewed. When a filter is chosen from the list, a brief text description of its effect appears below the Available window.

The complete list of Lumiere filter types, divided into nine categories, is shown in Figure 4.32. Editing options vary with the type of filter selected. Some examples are described below and shown in Figure 4.33.

Adobe Premiere includes 58 movie and still-image filters (Figure 4.34) to distort, blur, sharpen, smooth, texture, and colour images. Not surprisingly, many of these are reminiscent of Adobe Photoshop’s filter set; in fact many of the third party filters which work with Photoshop can be copied from Photoshop’s Plugins directory to Premiere’s Plugins directory, to extend the range of possibilities even further.

Premiere’s Filters dialog box (Figure 4.35) is similar to that of Media Studio and Lumiere. Filters to be applied to a selected clip are selected from the Available list and added to the Current list using the Add button. Some filter types have their own Settings dialog box, such as that shown in Figure 4.36 for the Ripple filter. Others do not and their effect can only be seen by previewing the clip after the filter has been applied. In the Construction window, clips with filters applied to them are displayed with a blue stripe along the top.

Editing in Drawing and Painting Applications

Creating Still Image Frames

For many video professionals and enthusiasts, videoediting is just that – editing video. The bulk of their work involves manipulating and editing video footage, with perhaps just the addition of the occasional title clip. For others – particularly those creating educational or promotional material – the use of drawing and painting applications gives access to (a) a much wider range of raw material which can be combined with video and audio to create what has become known as multimedia productions and (b) sophisticated still image editing techniques developed over many years. Still images produced using these applications can be interleaved with video material or shown alongside it, using a split screen technique. Some of the potential benefits are summarised below:

  Access to a much wider range of text effects than those provided by videoediting applications, as we saw earlier in this chapter

  Use of sophisticated vector drawing tools and techniques (e.g. blending, extrusion, enveloping) and special photoediting features such as channels, layers and complex paths

  Inclusion of 3D effects

  Preparation of ‘slides’ in a drawing application incorporating bulleted text, graphs, or tables, or design of charts using a combination of text, graphics and images

  Incorporation of drawings from clipart libraries in creating still frames, used ‘as is’ or edited and combined with other elements

  Access to Photo CD libraries of professionally shot still images

  Use of screen grabs, fractals or public domain images downloaded from the Internet

  Use of scanned 2D and 3D images and digital still images

  Selection of a frame from a video clip, for clean-up, editing or embellishing before reinsertion into a project as a title background

Any of the stills created using the above sources and techniques can be displayed as a still video clip within a project, or as a still insert within a moving video screen. For example a documentary movie could include a series of still images displayed in one half of a split screen while a narrator talks about their contents in the other half of the screen.

Figures 4.37 to 4.41 show examples of the kind of ‘slides’ which can easily be imported from drawing and painting applications into a video project.

Media Studio comes bundled with its own bitmap editing application called Image Editor (Figure 4.42) which is adequate for basic image manipulation. Corel provides PHOTO-PAINT, a much more sophisticated image editor, with Lumiere. Premiere does not provide an image editor but, as you would expect, works well with its market-leading photoediting application Photoshop. For the production of line drawings, any of the leading applications like Freehand, Illustrator or CorelDRAW are rich in features, while applications like RayDream Studio or Extreme 3D can be used to create impressive 3D graphics. In addition, specialist applications like Bryce and Poser are capable of producing quite amazing landscapes and posed human figures.

There is not sufficient space in this book to describe the remarkable range of graphic effects which can be produced using a combination of the above applications. Readers interested in learning more may wish to refer to previous books by the author – Digital Graphic Design and Digital Colour in Graphic Design – as well as other titles listed in the bibliography.

Editing through the Use of Scripts

The use of applications like PHOTO-PAINT and Painter in video production is not limited to the creation of still images; both applications provide the means of applying special effects to a series of frames within a video clip without the need to address each frame in turn. They do this through the use of scripts – also called macros in some applications. Scripts automate a series of actions which can be repeated on the same image or on several different images. A script is a recording of the series of editing actions which can be saved to disk for retrieval at any time. In PHOTO-PAINT the recording of such a series of actions and the resulting script can be edited in Corel’s SCRIPT Editor and played back using the tape deck controls and commands in a Recorder window which looks and behaves much like a tape recorder (Figure 4.43a). Activating this script (Figure 4.43b) opens a box (c) into which text is entered using the script language protocols. Clicking on OK then triggers a series of image editing actions which create a video clip ten frames in length depicting the text – the single word INFERNO, in this case – being consumed by flickering flames. The clip is automatically displayed in a window (Figure 4.43d).

In still image work, scripts save time when performing standard operations such as resampling images, using masks to make selections or making global adjustments to different images. Using the record button in the Recorder window, almost all keyboard, toolbar, toolbox, menu, and mouse actions can be recorded. As the recording takes place, the actions are translated into command statements which are numbered chronologically in the command list. Playing or running a script applies the recorded actions to the active image, like the text string in the example below.

PHOTO-PAINT provides a number of prerecorded scripts which can be applied to still images or video clips (Figure 4.44). The majority of these create text effects but examples of scripts which can be applied to video clips are included, for example:

  Movie Builder – builds a movie using sequentially numbered still image files stored in the same directory (e.g. imageOI .tif, image02.tif, etc.)

  Movie Fader – produces a fade out effect over a series of movie frames

Once created, scripts can be edited like any text file, using cut, copy or paste to remove, modify or add actions to the same script or to a new one.

Painter employs a similar VCR deck metaphor to that of PHOTO-PAINT for recording and playing back scripts. The deck is activated by clicking Script in the Objects palette (Figure 4.45). Used correctly, a script can provide a powerful educational tool. For example, playing back the script of an art project – the rendering of the Mona Lisa in Figure 4.46 – demonstrates the step-by-step process used by the artist to create it. As the script plays, the artwork evolves and develops just as it did when it was first created. Controls on the deck can be used to pause, replay or step forwards or backwards through the development of the piece.

A Painter script can also be used to create a video clip of the process represented by the script; the editing actions described within the script become frames within a new video clip by carrying out the following steps:

  The recorded session is first loaded via the Objects: Scripts window

  A new Painter image is opened at the size at which the video clip is to be created

  Script Options is selected from the Script dropdown menu, opening the dialog box in Figure 4.47a

  Clicking Save Frames on Playback directs Painter to create a video frame from each process step. The frame rate is also set from this box (the lower the number entered, the higher the frame rate)

  Clicking the Play button in the Script palette opens the dialog box in (b) asking for the parameters for a new frame stack (a frame stack is a video clip separated into its individual frames – see more on this subject later)

  After the frame stack parameters have been set, another dialog box opens, asking for a name and a Save destination for the new frame stack. Clicking OK in this box, after entering a file name, causes Painter to play the script back into a new frame stack and opens the Frame Stacks dialog box (c)

  Finally, the frame stack has to be converted to a video format. Clicking on File/Save As opens the dialog box in Figure 4.48 where the AVI option and frame rate can be selected before clicking on OK to save the Painter session as an AVI file

Painter scripts can also be used to apply special effects to video clips using just a single command. Figure 4.49 shows an example of one of Painter’s libraries, in which a thumbnail represents each of the effects. The procedure for applying an effect to a video clip is as follows:

  The video clip to which the effect is to be applied is first opened and converted to frame stack format (Figure 4.50a)

  Apply Script to Movie is selected from Painter’s Movie menu

  Painter opens a dialog box which lists the scripts in the current library by name (Figure 4.50b)

  When Playback is clicked in the Recorded Scripts dialog box, the script – in this case Corduroy Effect – is applied to each of the frames in the frame stack in turn (Figure 4.50c). The time taken depends on the number of frames in the stack

  Finally the frame stack can be saved, once again, as an AVI video clip

Painter provides several libraries of special effects scripts. Figure 4.51 shows just a few examples.

Rotoscoping, Compositing, Cloning and Tracing

Rotoscoping is the process of painting and applying effects to an existing video clip – for example, a captured video segment – or compositing a portion of the images from one clip with the images of another. This is a time-consuming process, as 25 frames have to be edited for each second of video, but it provides an enormous degree of control as all the tools and techniques of image editing applications like PHOTO-PAINT or Painter can be applied to the task. In some cases, where the changes are repetitive from frame to frame, a script can be created to apply changes made to the first frame, to successive frames.

Figure 4.52a shows an example of a frame from a clip displaying an empty, Disney-like, paved street scene. After opening the clip in PHOTO-PAINT, the application’s paintbrush and object manipulation tools were used to add a figure to the scene (Figure 4.52b). The figure was saved as an object, placed in the next frame (Figure 4.52c), and edited to create the effect of the figure striding along the street. Painting the figure’s shadow on the paving shows how this can help integrate the figure into the scene.

Rotoscoping is often used to put the action of a person or an object filmed in one place on a background filmed in another, or to remove an unwanted element from a video clip. Figure 4.53a shows a frame from a videoclip depicting a helicopter ambulance flying over a stretch of water. After creating a feathered mask to protect the helicopter image, a second image, of a rocky terrain was opened in PHOTO-PAINT and sized to match the resolution and frame size of the first image. The terrain image was then copied to the clipboard and PHOTO-PAINT’s Pasfe Inside command was used to paste the terrain ‘beneath’ the helicopter (Figure 4.53b). The same technique can be used to produce a wide range of effects; Figure 4.53c shows a mountain valley scene pasted beneath the helicopter.

Figure 4.54 shows how rotoscoping can be used to ‘clean up’ the image on a video clip. A frame from the original clip is shown in Figure 4.54a. Using PHOTO-PAINT’s cloning tool, the area below the white symbol in the top left corner of the frame was selected and the Clone tool was used to paint out the symbol (Figure 4.54b). Next, the area to the top right corner of the frame was selected and the Clone tool was used to paint out the face behind the main figure (Figure 4.54c).

Before a clip is opened in PHOTO-PAINT, the Partial Load Movie preview dialog box shown in Figure 4.55 appears. The slider or the VCR-like controls can be used to preview the clip and the section to be opened for editing can be selected either by dragging the two markers which can be seen at the ends of the slider, or by entering frame numbers in the boxes provided. This can reduce the memory problems which can arise when a large movie file is opened.

When opening a video clip in Painter for editing, it is also more efficient to open just the segment of the clip to be edited. When a clip is opened in Painter, it is automatically converted to a frame stack. Framestacks are uncompressed, so, for example, a 1 Mb AVI movie can easily become a 20 Mb framestack. If only the first 20 seconds of a 3-minute clip are to be edited, then these 20 seconds can be separated off in a videoediting application, edited in Painter and then returned to the original clip.

Painter can also be used to composite two video clips together into one – for example, to composite a foreground action against a new background – using the following procedure:

  First the foreground action must be masked in each frame by creating a selection. Because the foreground image continues to move, the selection in each frame must be different. (When the background is uniform – all white, for example – Painter’s automatic selection and script features can be used to mask each frame)

  When each frame of the foreground is properly protected by a selection, the video clip to be used as a background is opened (If the background is static, a single image can be used instead of a video clip)

  With the background clip active. Set Movie Clone Source is selected from the Movie menu

  With the foreground movie active, the cloning brush is selected from the Brushes palette. Painting in the foreground movie now fills in the background from the background clip

  After finishing the first frame, clicking the Step Forward button automatically advances the foreground and clone source clips by one frame. The clips stay synchronized as work proceeds

Figure 4.56 illustrates the technique. Figure 4.56a shows the first frame of a clip showing an African elephant doing what comes naturally in its natural habitat. The dotted line around the elephant shows where a selection mask has been placed around it. Figure 4.54b shows the first frame of a clip displaying a fantasy Arctic landscape. After setting the Arctic clip as the clone source, inverting the selection in the first clip and choosing the Straight doners from Painter’s Brushes/Goners menu, painting into the inverted selection produced the result shown in Figure 4.56c, transporting the elephant, dung and all, to a much less natural habitat!

Straight cloning reproduces exactly what appears in the source image. Painter also includes a number of other cloning tools which can be used to produce more artistic cloning effects. Figure 4.57 shows an example. A background image of a rose is shown in Figure 4.57a and a selected image from the first frame of a video clip is shown in Figure 4.57b. After inverting the selection of the girl and choosing the image of the rose as the clone source, the doners option was again selected from Painter’s Brushes palette, but this time the Soft doner option was selected, so that painting into the background created a ‘soft focus’ variation of the rose image. Figure 4.57d used the Impressionist variant of the doner, producing a background version of the rose image painted with bold dabs of paint, reminiscent of the Impressionist style.

Finally, for those with infinite patience wishing to produce a video with a unique, hand-finished look, Painter even allows tracing of video frames! Figures 4.58a and 4.58b show the original clip and frame stack. 4.58c and 4.58d show the result of tracing just the parrot into a new clip using different drawing styles for each frame.

Summary

In this chapter we have seen that video editing applications offer the designer a wide range of creative possibilities for the addition of titles, transitions and special effects to integrate and enhance raw video footage. For the designer experienced in the use of still image editing applications like Photoshop, Painter, PHOTO-PAINT, Illustrator or CorelDRAW, for example, the range of possibilities is even wider.

Title screens can be created using any of thousands of different typefaces and styles and a whole battery of type manipulation effects. Access can be gained to a much wider range of raw material such as professional quality clipart, high resolution PhotoCD images, scanned images and photographs. Still images can be used ‘as is’ or can be precisely edited before being inserted between video clips, with accompanying voice over, or can be combined with video using a split screen technique. Application of scripts to still images or video clips provides consistent and repeatable control over implementing frequently used changes.

Already in the field of still image editing there has been a significant convergence between bitmap and vector drawing applications which has allowed the designer to move effortlessly between them to combine their various features. We are beginning to see similar convergence between still and video application – particularly those developed by the same vendor. The effect of this will be to make it even easier in the future to combine the best of still image editing with the best of video editing to produce eye-catching multimedia combinations.