1 .TH P64 1 "14 June 1993"
2 .UC 4
3 .SH NAME
4 p64 \- P64 compression and decompression
5 .SH SYNOPSIS
6 .B
7 p64 [-d [-c]] [-NTSC] [-CIF] [-QCIF]
8 .B
9 [-a StartNumber] [-b EndNumber]
10 .B
11 [-f FrameRate] [-k FrameskipNumber] [-o] [-p]
12 .B
13 [-i MCSearchLimit] [-q Quantization] [-v] [-y]
14 .B
15 [-r Target Rate] [-x Target Filesize]
16 .B
17 [-s StreamFile] [-z ComponentFileSuffix i]
18 .B
19 ComponentFilePrefix1 [ComponentFilePrefix2 ComponentFilePrefix3]
20 .br
21 .SH DESCRIPTION
22 .I p64
23 is a telecommunication image sequence compression/decompression
24 program that performs CCITT H.261 encoding and decoding.
25 .PP
26 .SH OPTIONS
27 .TP
28 .B -NTSC
29 indicates that the source frame size for the Luminance component is
30 352x240 and for the Chrominance components is 176x120. This is the
31 default size.
32 .TP
33 .B -CIF
34 indicates that the source frame size for the Luminance component is
35 352x288 and for the Chrominance components is 176x144.
36 .TP
37 .B -QCIF
38 indicates that the source frame size for the Luminance component is
39 176x144 and for the Chrominance components is 88x72.
40 .TP
41 .B -a
42 denotes the following value to be the start of the filename index
43 [inclusive]. Defaults to 0.
44 .TP
45 .B -b
46 denotes the following value to be the end of the filename index
47 [inclusive]. Defaults to 0.
48 .TP
49 .B -c
50 for the decoder, it indicates to save to a CIF style file. Always
51 compatible with QCIF or NTSC modes.
52 .TP
53 .B -d
54 enables the decoder.
55 The only options useful for the decoder are the filename specifications;
56 the rest of the information is used for encoding.
57 .TP
58 .B -f
59 is the FrameRate, or the number of frames per second. The default
60 number is 30 (frames/second), regardless of
61 .B -PAL
62 or
63 .B -NTSC
64 mode.
65 .TP
66 .B -i
67 is the diameter of the motion estimation search. The diameter must be
68 inclusively between 1 and 31, designating a region from interframe
69 only (0,0) search, to full search (+- 15, +- 15). (Default 15).
70 .TP
71 .B -k
72 is the FrameSkip index, representing the spacing between each frame.
73 A value of 1 indicates that every frame is to be transmitted. A value
74 of 2 indicates every other frame is to be transmitted. In general, the
75 number of transmitted frames per second is the FrameRate divided by
76 FrameSkip.
77 .TP
78 .B -o
79 signals that the program interpreter will read the control algorithms
80 from the standard input.
81 .TP
82 .B -p
83 enables parity checking for the frames.
84 (Disabled under normal compilation for 1992
85 CCITT specifications.) This checks the validity of
86 the reference frame stores on the decoder end. While the parity is
87 correct for this system's encoder and decoder, some parity errors (on
88 the least significant bits) can occur in
89 .I normal
90 operation because of unmatched inverse DCT's between arbitrary encoders and
91 decoders.
92 .TP
93 .B -q
94 gives a value for the quantization
95 not in the presence of rate control,
96 which automatically changes the quantization values.
97 If rate control is specified, this parameter gives a value for the
98 initial quantization of the first frame (which the program usually
99 takes an educated guess at).
100 .TP
101 .B -r
102 specified a rate for the coded stream in bits per second. If this is
103 enabled, a buffer model is used to limit the size of the coding
104 stream.
105 .TP
106 .B -s
107 specifies the filename to store the coded image. If unspecified it
108 defaults to
109 .B ComponentFilePrefix0.p64.
110 .TP
111 .B -v
112 designates that quantization decisions are to be written to standard
113 output.
114 .TP
115 .B -x
116 gives a target filesize for the compressed stream. This overrides the
117 rate option
118 .B -r
119 , if specified. Note that the filesize will
120 generally be larger than the target filesize because of remaining bits
121 in the buffer.
122 .TP
123 .B -y
124 enables the double-precision floating point Reference DCT. The default
125 is the Chen DCT.
126 .TP
127 .B -z
128 denotes the component file suffixes in sequential order. For example
129 .B
130 -z .y.clr -z .u.clr -z .v.clr
131 indicates that the luminance and the two chrominance files
132 end with a suffix of
133 .B
134 \&.y.clr, \&.u.clr,
135 and
136 .B \&.v.clr,
137 in that order. If unspecified, the suffixes
138 default to
139 .B
140 \&.Y, \*.U,\fR
141 and
142 .B \&.V.
143 .TP
144 .B ComponentFilePrefix\fIn\fR
145 must be specified. In general, the other component files share the
146 same prefix and will default to
147 .B ComponentFilePrefix1
148 if not explicitly specified. However, in some cases the prefix is what
149 changes in the file and we can specify this individually.
150 .PP
151 .SH EXAMPLES
152 To encode a set of raw raster scan YCbCr (also known as digital YUV) files with
153 dimension 352x240 Y, 176x120 U,V,
154 with labels,
155 .B
156 hello0.Y hello0.U hello0.V,
157 .B
158 hello1.Y hello1.U hello1.V,
159 .B
160 \&...,
161 .B
162 hello12.Y hello12.U hello12.V
163 through the p64 encoder, type
164
165 .br
166 .B
167 p64 -a 0 -b 12 hello -s hello.p64
168 .br
169
170 This creates the output file
171 .B hello.p64.
172 In order to decode such a file into a sequence of image files with
173 prefix
174 .B goodbye,
175 type
176
177 .br
178 .B
179 p64 -d -s hello.p64 goodbye
180 .br
181
182 The output will be placed in the files
183 .B
184 goodbye0.Y goodbye0.U goodbye0.V,
185 .B
186 goodbye1.Y goodbye1.U goodbye1.V,
187 .B
188 \&...,
189 .B
190 goodbye12.Y goodbye12.U goodbye12.V.
191 These image sequences can be displayed by the
192 .I cv
193 program.
194 The image sequences can also be converted to ppm and back through
195 the programs
196 .I cyuv2ppm
197 and
198 .I ppm2cyuv
199 Those utility programs available by anonymous ftp from
200 .I havefun.stanford.edu:pub/cv/CVv1.2.1.tar.Z.
201 There are more options within an internal program interpreter.
202 Please see the accompanying documentation in
203 .I doc.ps
204 for more details.
205 .PP
206 .SH FTP
207 .I p64
208 is available by anonymous ftp from
209 .I havefun.stanford.edu:pub/p64/P64v1.2.tar.Z.
210 .PP
211 .SH BUGS
212 Somewhat slower than many commercial implementations.
213 Please inform the author at achung@cs.stanford.edu if you have
214 such a sequence available or if any bugs are found.
215 .PP
216 .SH AUTHOR
217 .PP
218 Andy Hung
This page was automatically generated by the
LXR engine.
Visit the LXR main site for more
information.