Fields, Pulldown, and Pixel Aspect Ratio
One surprise for the novice is that moving images are often
not made up of whole frames containing square pixels
like stills. A video frame, and in particular one shot for
broadcast, is often interlaced into two fi elds, and its pixels
are stored non-square, for the purpose of faster and more
effi cient capture and delivery.
Fields combine two frames into one by interlacing them
together, vertically alternating one horizontal line of pixels
from the fi rst with one from the second. The result is half
the image detail but twice the motion detail. Figure 1.15
shows this principle in action.
If you’re doing any compositing, transformation, paint/
masking, or distortion —pretty much anything beyond basic
color correction—it’s best to match the Separate Fields setting
to that of the footage, causing After Effects to recognize
the interlace frame as two separate frames of video.
Pulldown uses fi elds to run 24 fps fi lm footage smoothly
at 29.97 fps by repeating one fi eld every fi ve frames. This
creates a pattern that After Effects can accurately guess if
there is suffi cient motion in the fi rst few frames of the footage.
If not, the backup option (which still works) is trialand-
error, trying each initial pattern listed under Remove
Pulldown until the fi eld artifacts disappear in a 23.976 fps
comp. There are two basic types of pulldown (3:2 and 24
Pa), each with fi ve potential initial patterns.
Pixel aspect ratio (PAR) is another compromise intended
to maximize image detail while minimizing frame size. The
pixels in the image are displayed nonsquare on the broadcast
monitor, with extra detail on one axis compensating
for its lack on the other.
Your computer monitor, of course, displays square pixels,
so any clip with a non-square PAR will look odd if displayed
without compensating for the difference. Therefore, After
Effects includes a toggle below the viewer panels to stretch
the footage so that its proportions look correct (Figure 1.17)
although the footage or composition itself isn’t changed.
With some digital formats such as DV, fi eld order and
pixel aspect are standardized and set automatically in After
Effects. With other formats, it’s best to know the correct
fi eld order and pixel aspect as specifi ed by the camera or
software that generated the image.
Source Formats
After Effects is capable of importing and exporting a wide
array of footage formats, yet only a small subset of these
recur regularly in production. Table 1.1 contains a rundown
of common raster image formats and some advantages
and disadvantages of each.
Which formats will you use most? Probably TIFF or DPX
for source, and JPEG (with a Quality setting of 7 or higher)
for temporary storage when fi le space is at a premium.
TIFF offers lossless LZW compression, giving it an advantage
over Adobe Photoshop, especially when you consider
that TIFF can even store multiple layers, each with its own
transparency. Other formats with lossless compression,
such as TGA, don’t support multiple bit-depths and layers
like TIFF does. PNG is more limited and slower, but the fi le
sizes are smaller.
For fi lm and computer graphics, it is normal to pass
around CIN and DPX fi les (essentially the same format)
and EXR, designed (and open-sourced) by ILM specifi cally
to handle HDR renders with multiple channels of data
(and these can be customized to contain useful information
such as Z depth and motion data). More on these
formats is found in Chapters 11 and 12, which also include
information on working with Camera Raw CRW images.
One surprise for the novice is that moving images are often
not made up of whole frames containing square pixels
like stills. A video frame, and in particular one shot for
broadcast, is often interlaced into two fi elds, and its pixels
are stored non-square, for the purpose of faster and more
effi cient capture and delivery.
Fields combine two frames into one by interlacing them
together, vertically alternating one horizontal line of pixels
from the fi rst with one from the second. The result is half
the image detail but twice the motion detail. Figure 1.15
shows this principle in action.
If you’re doing any compositing, transformation, paint/
masking, or distortion —pretty much anything beyond basic
color correction—it’s best to match the Separate Fields setting
to that of the footage, causing After Effects to recognize
the interlace frame as two separate frames of video.
Pulldown uses fi elds to run 24 fps fi lm footage smoothly
at 29.97 fps by repeating one fi eld every fi ve frames. This
creates a pattern that After Effects can accurately guess if
there is suffi cient motion in the fi rst few frames of the footage.
If not, the backup option (which still works) is trialand-
error, trying each initial pattern listed under Remove
Pulldown until the fi eld artifacts disappear in a 23.976 fps
comp. There are two basic types of pulldown (3:2 and 24
Pa), each with fi ve potential initial patterns.
Pixel aspect ratio (PAR) is another compromise intended
to maximize image detail while minimizing frame size. The
pixels in the image are displayed nonsquare on the broadcast
monitor, with extra detail on one axis compensating
for its lack on the other.
Your computer monitor, of course, displays square pixels,
so any clip with a non-square PAR will look odd if displayed
without compensating for the difference. Therefore, After
Effects includes a toggle below the viewer panels to stretch
the footage so that its proportions look correct (Figure 1.17)
although the footage or composition itself isn’t changed.
With some digital formats such as DV, fi eld order and
pixel aspect are standardized and set automatically in After
Effects. With other formats, it’s best to know the correct
fi eld order and pixel aspect as specifi ed by the camera or
software that generated the image.
Source Formats
After Effects is capable of importing and exporting a wide
array of footage formats, yet only a small subset of these
recur regularly in production. Table 1.1 contains a rundown
of common raster image formats and some advantages
and disadvantages of each.
Which formats will you use most? Probably TIFF or DPX
for source, and JPEG (with a Quality setting of 7 or higher)
for temporary storage when fi le space is at a premium.
TIFF offers lossless LZW compression, giving it an advantage
over Adobe Photoshop, especially when you consider
that TIFF can even store multiple layers, each with its own
transparency. Other formats with lossless compression,
such as TGA, don’t support multiple bit-depths and layers
like TIFF does. PNG is more limited and slower, but the fi le
sizes are smaller.
For fi lm and computer graphics, it is normal to pass
around CIN and DPX fi les (essentially the same format)
and EXR, designed (and open-sourced) by ILM specifi cally
to handle HDR renders with multiple channels of data
(and these can be customized to contain useful information
such as Z depth and motion data). More on these
formats is found in Chapters 11 and 12, which also include
information on working with Camera Raw CRW images.
No comments:
Post a Comment