Table of contents:
Introduction | Introduction | Defining
Codec | Compression | Windows Media
| Real System 8 | QuickTime5 | Divx
|
Assignments | Additional References
| Lesson 3.2
Few media are as engaging as video. Full motion video has become a daily part of modern life (who's watched a movie or television in the last week?) Most people now take multimedia, and the equipment you need to enjoy it (a video receptor and an audio playback device), for granted. But even though almost everyone watches video, creating quality video is still a complicated and often expensive process. As our text quotes,
"Full motion video on personal computers changes everything. It is like turning a ten-speed bike into a Harley-Davidson."--David Bunnell
The capture and compression process for video is slow and cumbersome, and raw, captured video is huge. How huge? Ten seconds of raw, uncompressed NTSC video (which is the standard for television) will fill as much as 300 MB of storage space.
| Video | An electrical signal containing timing (synchronization), luminance (intensity), and often chrominance (color) information that, when displayed on an appropriate device, gives a visual image or representation of the original image sequences. |
| Non-Linear Editing | The project can be assembled in any order from beginning to end, and changes can be made in the project anywhere at any time. |
| Frame Rate | The number of frames per second (30 fps is optimum, 15 fps is acceptable) |
| Color depth | Digitized video is really a series of still grpahic bitmaps displayed according to a preset frame rate. So, the quality of the video depends on the color quality (related to the number of colors) for each graphic bitmap. 8-bit color provides 256 colors, 16-bit color provides more than 64,000 colors, and 24-bit color provides more than 16 million colors. |
| Frame Size | The size of the frame, i.e., 320 x 240, 640 x 480, and so on. |
| Aspect Ratio | width:height. Common aspect ratios are 4:3 and 16:9 |
| Codec | As with audio, a codec is a compression/decompression program that provides a set of standards by which the player might interpret the video graphics. |
| Keyframes | Frame where the scene changes. Every video has at least one (frame 1). |
| Transition | Change from one scene to another. |
| Rendering/Previewing | Coding the video and displaying the result. |
| Chrominance | The color portion of a video signal carrying the saturation and tint (hue) information for any given point in the image. |
| Luminance | The information about brightness, darkness and contrast |
| Scaling | Sizing the video for a window. |
Before beginning this Unit, please read Chapter 12 in your text. There is more in there than we are going to cover in this unit, but knowing more will not hurt you. Knowing less will, though.
Analog video is generally videotape, and it is linear. That is, it has a beginning, a middle and an end. If you want to edit it, you need to rewind, pause, and then fast forward through the tape to display the desired frames. Smaller files, harder to work with.
Digital video, created and edited on digital media, is non-linear. That is, you can access any part of it without fast forwarding or rewinding (random access). Editing digitized video is analogous to cutting and pasting in a word processor. Easy to work with, large files.
There will be more on editing video in Lessons 3.3 and 3.4.
"Codec" is an abbreviation for compression/decompression. A codec can be either a software application or a piece of hardware that processes video through complex algorithms, which compress the file and then decompress it for playback. Unlike other kinds of file-compression packages that require you to decompress a file before viewing, video codecs decompress the video on the fly, allowing the client to view the file from its compressed original.
Some codecs are more appropriate for certain kinds of work than others. Which codecs are available to you when editing digital video in your editor depends on your system and capture card. This choice is further constrained by the editing mode you choose in in the editor. For example, in Windows the video codecs you can choose will be different if you choose Video for Windows as your Editing Mode than if you choose Quicktime. When you export digital video, the available codecs are determined by the the settings available in your editor.
There are two main kinds of compression, lossy, and loss-less. In loss-less compression the original data is compressed, and when de-compressed, the data has not changed. Loss-less compression is often applied to still pictures. Lossy compression includes most video compression formats, such as JPEG, MPEG-1/2/4. Lossy compression means that when the digital data is compresssed and de-compressed, it is no longer the same as the original (usually denoting a loss of quality). Lossy compression actually eliminates some of the data in the image and therfore provides greater compression ratios than loss-less compression. The greater the compression ratio is, the worse the decompressed image is. The trade-off is file size vs. image quality. Lossy compression is often applied to video because some drop in the quality is no noticeable in moving images.
Video codecs work in two ways - using temporal and spatial compression.
This method of compression looks for information that is not necessary for continuity to the human eye or ear (remember that videotape plays back sound as well as pictures). It looks at the video information on a frame-by-frame basis for changes between frames. For example, if you're working with video of a talking head (a clip of a person sitting or standing with little motion), there's a lot of redundant information in the recording. The background rarely changes, and most of the motion involved is simple head movements and the movement of the area around the mouth. The compression algorithm compares the first frame (known as a key frame) with the next (called a delta frame) to find anything that changes. After the key frame, it only keeps the information that does change, thus deleting a large portion of your file. It does this for each frame until it reaches the end of the file. If there is a scene change, it tags the first frame of the new scene as the next key frame and continues comparing the following frames with this new key frame. As the number of key frames increases, so does the file size.
Spatial compression uses a different method to delete information that is common to the entire file or an entire sequence within the file. It also looks for redundant information, but instead of specifying each pixel in an area, it defines that area using coordinates. Both of these compression methods reduce the overall file size. If this is not sufficient, one can make a larger reduction in file size by reducing colors, frame rate (that is, how many frames of video go by in a given second), and audio quality. Depending on the degree of changes one makes in each of these areas, the final output can vary greatly in quality.
The quality, frame rate, and frame size of Microsoft® Windows Media™-based content is largely determined by the codec (compressor/decompressor) you select when creating content. The fact that Windows Media Technologies is codec-independent means that you have a choice of codecs and settings for compressing audio and video content.
| Codec | When to use |
|---|---|
|
Microsoft Windows Media Audio version 7.0 |
Encoding most audio streams. It provides unmatched audio quality for Internet narrowband streaming and high-fidelity downloadable music. Improvements in quality over previous codecs were made at all data rates from 16 kilobits per second (Kbps) to 192 Kbps. This latest Windows Media Audio codec supports content encoded with Windows Media Audio version 2.0. |
|
Sipro Labs ACELP |
Encoding low-bit-rate voice content. This codec comes with several audio formats depending on the network bandwidth you choose. |
|
Microsoft Windows Media Video version 7.0 |
Encoding most video streams. It supports a wide variety of network bandwidths and enhances video quality for broadband Internet users. It supports multiple-bit-rate profiles and delivers TV-quality compression at 700 Kbps. |
|
Microsoft MPEG-4 version 3.0 |
Encoding video with Windows Media Encoder version 4.0. Although all codecs mentioned here are included with Windows Media Encoder 7, there may be the occasion when you must use the version 4.0 codec for compatibility with older versions of Windows Media Player, for example. |
|
ISO MPEG-4 video codec version 1.0
|
Based on the International Standards Organization (ISO) MPEG-4 standard. You can use this codec to encode content produced by many consumer electronics devices, such as digital video cameras and cellular phones. |
|
Microsoft Windows Media Screen version 7.0 (now version 8) |
Encoding screen capture video. This codec enables lossless streaming of computer screen images at data rates as low as 15 kbps. You can use it for on-demand training, software demonstrations, and helpdesk applications. It is well suited for capturing screens in productivity applications, such as Microsoft Word, Excel, and Visual Studio®. Windows Media Screen codec is fully optimized for both streaming and download scenarios and makes use of digital rights management (DRM) technology to encrypt the encoded screen content. It is included with Windows Media Player 7. |
Real Media is a streaming format based around MPEG 2 formats that has evolved quickly. One of the most popular web-only formats, Real Video 8 gives you both quality and control over your work. Its compressor is state of the art, including Variable Bit Rate compression. VBR helps to enhance picture quality while removing repeated data. This means a higher quality picture can fit into the same datarate. Real Media seems to degrade better at lower qualities and tends to offer better pictures quality at lower bitrates. Real offers a 'free' encoder, but to get full encoder functionality and the ability to encode at very high quality, you have to pay a license fee to Real. Also, the 'free' version of the player doesn't work as well as the 'Plus' player, meaning that Real has crippled the decoding software inside the software.
QuickTime is popular because it runs on both Windows and Mac platforms. It uses lossy compression coding and can achieve ratios ranging from 5:1 to 25:1. Based on the Sorenson Video 3 codec, QuickTime 5 promises to deliver higher-quality content. MPEG-1 streaming support is already built-in to QT 5, and Apple is currently working on MPEG-2 support. QuickTime 5 also supports a wider range of audio, video, and graphics formats, including Flash 5 and MPEG-1 plus proprietary formats such as iPix, Pulse, SHOUTcast, and On2. This eliminates the need for separate media players.
The later versions of Quicktime 3, 4, and 5 also include a QuickTime video compression codec that is very similiar to the MPEG 4 codecs available. To encode video, you are required to pay a license fee to Apple for Quicktime Pro.
Video compressors in QuickTime:
- H.261
- H.263
- Animation
- Apple BMP
- Apple Video
- Cinepak Component video
- DV NTSC and PAL
- Graphics
- Microsoft OLE
- Microsoft Video 1
- Motion JPEG A and B
- Photo JPEG
- Planar RGB
- Sorenson Video 1, 2, and 3
DivX is the name given to a video codec (a piece of software encoding and decoding video) and is based on the MPEG-4 compression format. MPEG-4 is a new standard of video compression that is both high quality and low bitrate. They are usually only a fraction (around 15%) of the size of a standard DVD, even at 640x480 resolutions, making them the best home video format thus far. They only take half the time to encode, and yet at the same time is smaller in size than MPEG-1 - due to their incredible compression technology - some have even called MPEG-4 the "MP3 of the video world". Quality ranges from net-streaming quality to DVD and better.
The only software problem with DivX is that the encoder tends to be somewhat unstable and can cause crashes. The player seems to work flawlessly. Legally, DivX is in somewhat more uncertain waters. Movie Studios *hate* DivX because video pirates can compress an entire DVD's worth of video to fit on a CD with only minimal degradation of quality.
There are also other version of the DivX codec, available for different operating systems :
- Go to Luke's Video Codec Comparison and describe the differences between uncompressed video and three other codecs. Email me your analysis. (if that is no longer available, email me and I'll send you the files.)
- Find two sites that offer video, including one with streaming video and one with downloaded video (CNN.com is always a good choice for video). Document the experiences that you have with each type of video, and relate all the factors that play into your Web video experience (your connection rate, time of day, computer configuration, etc.). Email your report.
When you have completed and emailed these assignments, please go on to Lesson 3.2.
Why Video is Crucially Different from Graphics