cwebp.1 (14132B)
1 .\" Hey, EMACS: -*- nroff -*- 2 .TH CWEBP 1 "April 10, 2025" 3 .SH NAME 4 cwebp \- compress an image file to a WebP file 5 .SH SYNOPSIS 6 .B cwebp 7 .RI [ options ] " input_file \-o output_file.webp 8 .br 9 .SH DESCRIPTION 10 This manual page documents the 11 .B cwebp 12 command. 13 .PP 14 \fBcwebp\fP compresses an image using the WebP format. 15 Input format can be either PNG, JPEG, TIFF, WebP or raw Y'CbCr samples. 16 Note: Animated PNG and WebP files are not supported. 17 .SH OPTIONS 18 The basic options are: 19 .TP 20 .BI \-o " string 21 Specify the name of the output WebP file. If omitted, \fBcwebp\fP will 22 perform compression but only report statistics. 23 Using "\-" as output name will direct output to 'stdout'. 24 .TP 25 .BI \-\- " string 26 Explicitly specify the input file. This option is useful if the input 27 file starts with a '\-' for instance. This option must appear \fBlast\fP. 28 Any other options afterward will be ignored. 29 .TP 30 .B \-h, \-help 31 A short usage summary. 32 .TP 33 .B \-H, \-longhelp 34 A summary of all the possible options. 35 .TP 36 .B \-version 37 Print the version number (as major.minor.revision) and exit. 38 .TP 39 .B \-lossless 40 Encode the image without any loss. For images with fully transparent area, 41 the invisible pixel values (R/G/B or Y/U/V) will be preserved only if the 42 \-exact option is used. 43 .TP 44 .BI \-near_lossless " int 45 Specify the level of near\-lossless image preprocessing. This option adjusts 46 pixel values to help compressibility, but has minimal impact on the visual 47 quality. It triggers lossless compression mode automatically. The range is 0 48 (maximum preprocessing) to 100 (no preprocessing, the default). The typical 49 value is around 60. Note that lossy with \fB\-q 100\fP can at times yield 50 better results. 51 .TP 52 .BI \-q " float 53 Specify the compression factor for RGB channels between 0 and 100. The default 54 is 75. 55 .br 56 In case of lossy compression (default), a small factor produces a smaller file 57 with lower quality. Best quality is achieved by using a value of 100. 58 .br 59 In case of lossless compression (specified by the \fB\-lossless\fP option), a 60 small factor enables faster compression speed, but produces a larger file. 61 Maximum compression is achieved by using a value of 100. 62 .TP 63 .BI \-z " int 64 Switch on \fBlossless\fP compression mode with the specified level between 0 65 and 9, with level 0 being the fastest, 9 being the slowest. Fast mode 66 produces larger file size than slower ones. A good default is \fB\-z 6\fP. 67 This option is actually a shortcut for some predefined settings for quality 68 and method. If options \fB\-q\fP or \fB\-m\fP are subsequently used, they will 69 invalidate the effect of this option. 70 .TP 71 .BI \-alpha_q " int 72 Specify the compression factor for alpha compression between 0 and 100. 73 Lossless compression of alpha is achieved using a value of 100, while the lower 74 values result in a lossy compression. The default is 100. 75 .TP 76 .BI \-preset " string 77 Specify a set of pre\-defined parameters to suit a particular type of 78 source material. Possible values are: \fBdefault\fP, \fBphoto\fP, 79 \fBpicture\fP, \fBdrawing\fP, \fBicon\fP, \fBtext\fP. Since 80 \fB\-preset\fP overwrites the other parameters' values (except the 81 \fB\-q\fP one), this option should preferably appear first in the 82 order of the arguments. 83 .TP 84 .BI \-m " int 85 Specify the compression method to use. This parameter controls the 86 trade off between encoding speed and the compressed file size and quality. 87 Possible values range from 0 to 6. Default value is 4. 88 When higher values are used, the encoder will spend more time inspecting 89 additional encoding possibilities and decide on the quality gain. 90 Lower value can result in faster processing time at the expense of 91 larger file size and lower compression quality. 92 .TP 93 .BI \-crop " x_position y_position width height 94 Crop the source to a rectangle with top\-left corner at coordinates 95 (\fBx_position\fP, \fBy_position\fP) and size \fBwidth\fP x \fBheight\fP. 96 This cropping area must be fully contained within the source rectangle. 97 Note: the cropping is applied \fIbefore\fP any scaling. 98 .TP 99 .BI \-resize " width height 100 Resize the source to a rectangle with size \fBwidth\fP x \fBheight\fP. 101 If either (but not both) of the \fBwidth\fP or \fBheight\fP parameters is 0, 102 the value will be calculated preserving the aspect\-ratio. Note: scaling 103 is applied \fIafter\fP cropping. 104 .TP 105 .BI \-resize_mode " string 106 Specify the behavior of the \fB\-resize\fP option. Possible values are: 107 \fBdown_only\fP, \fBup_only\fP, \fBalways\fP (default). \fBdown_only\fP will 108 use the values specified by \fB\-resize\fP if \fIeither\fP the input width or 109 height are larger than the given dimensions. Similarly, \fBup_only\fP will only 110 resize if \fIeither\fP the input width or height are smaller than the given 111 dimensions. 112 .TP 113 .B \-mt 114 Use multi\-threading for encoding, if possible. 115 .TP 116 .B \-low_memory 117 Reduce memory usage of lossy encoding by saving four times the compressed 118 size (typically). This will make the encoding slower and the output slightly 119 different in size and distortion. This flag is only effective for methods 120 3 and up, and is off by default. Note that leaving this flag off will have 121 some side effects on the bitstream: it forces certain bitstream features 122 like number of partitions (forced to 1). Note that a more detailed report 123 of bitstream size is printed by \fBcwebp\fP when using this option. 124 125 .SS LOSSY OPTIONS 126 These options are only effective when doing lossy encoding (the default, with 127 or without alpha). 128 129 .TP 130 .BI \-size " int 131 Specify a target size (in bytes) to try and reach for the compressed output. 132 The compressor will make several passes of partial encoding in order to get as 133 close as possible to this target. If both \fB\-size\fP and \fB\-psnr\fP 134 are used, \fB\-size\fP value will prevail. 135 .TP 136 .BI \-psnr " float 137 Specify a target PSNR (in dB) to try and reach for the compressed output. 138 The compressor will make several passes of partial encoding in order to get as 139 close as possible to this target. If both \fB\-size\fP and \fB\-psnr\fP 140 are used, \fB\-size\fP value will prevail. 141 .TP 142 .BI \-pass " int 143 Set a maximum number of passes to use during the dichotomy used by 144 options \fB\-size\fP or \fB\-psnr\fP. Maximum value is 10, default is 1. 145 If options \fB\-size\fP or \fB\-psnr\fP were used, but \fB\-pass\fP wasn't 146 specified, a default value of '6' passes will be used. If \fB\-pass\fP is 147 specified, but neither \fB-size\fP nor \fB-psnr\fP are, a target PSNR of 40dB 148 will be used. 149 .TP 150 .BI \-qrange " int int 151 Specifies the permissible interval for the quality factor. This is particularly 152 useful when using multi-pass (\fB\-size\fP or \fB\-psnr\fP options). 153 Default is 0 100. 154 If the quality factor is outside this range, it will be clamped. 155 If the minimum value must be less or equal to the maximum one. 156 .TP 157 .B \-af 158 Turns auto\-filter on. This algorithm will spend additional time optimizing 159 the filtering strength to reach a well\-balanced quality. 160 .TP 161 .B \-jpeg_like 162 Change the internal parameter mapping to better match the expected size 163 of JPEG compression. This flag will generally produce an output file of 164 similar size to its JPEG equivalent (for the same \fB\-q\fP setting), but 165 with less visual distortion. 166 167 .TP 168 Advanced options: 169 170 .TP 171 .BI \-f " int 172 Specify the strength of the deblocking filter, between 0 (no filtering) 173 and 100 (maximum filtering). A value of 0 will turn off any filtering. 174 Higher value will increase the strength of the filtering process applied 175 after decoding the picture. The higher the value the smoother the picture will 176 appear. Typical values are usually in the range of 20 to 50. 177 .TP 178 .BI \-sharpness " int 179 Specify the sharpness of the filtering (if used). 180 Range is 0 (sharpest) to 7 (least sharp). Default is 0. 181 .TP 182 .B \-strong 183 Use strong filtering (if filtering is being used thanks to the 184 \fB\-f\fP option). Strong filtering is on by default. 185 .TP 186 .B \-nostrong 187 Disable strong filtering (if filtering is being used thanks to the 188 \fB\-f\fP option) and use simple filtering instead. 189 .TP 190 .B \-sharp_yuv 191 Use more accurate and sharper RGB->YUV conversion. Note that this process is 192 slower than the default 'fast' RGB->YUV conversion. 193 .TP 194 .BI \-sns " int 195 Specify the amplitude of the spatial noise shaping. Spatial noise shaping 196 (or \fBsns\fP for short) refers to a general collection of built\-in algorithms 197 used to decide which area of the picture should use relatively less bits, 198 and where else to better transfer these bits. The possible range goes from 199 0 (algorithm is off) to 100 (the maximal effect). The default value is 50. 200 .TP 201 .BI \-segments " int 202 Change the number of partitions to use during the segmentation of the 203 sns algorithm. Segments should be in range 1 to 4. Default value is 4. 204 This option has no effect for methods 3 and up, unless \fB\-low_memory\fP 205 is used. 206 .TP 207 .BI \-partition_limit " int 208 Degrade quality by limiting the number of bits used by some macroblocks. 209 Range is 0 (no degradation, the default) to 100 (full degradation). 210 Useful values are usually around 30\-70 for moderately large images. 211 In the VP8 format, the so\-called control partition has a limit of 512k and 212 is used to store the following information: whether the macroblock is skipped, 213 which segment it belongs to, whether it is coded as intra 4x4 or intra 16x16 214 mode, and finally the prediction modes to use for each of the sub\-blocks. 215 For a very large image, 512k only leaves room for a few bits per 16x16 216 macroblock. 217 The absolute minimum is 4 bits per macroblock. Skip, segment, and mode 218 information can use up almost all these 4 bits (although the case is unlikely), 219 which is problematic for very large images. The partition_limit factor controls 220 how frequently the most bit\-costly mode (intra 4x4) will be used. This is 221 useful in case the 512k limit is reached and the following message is displayed: 222 \fIError code: 6 (PARTITION0_OVERFLOW: Partition #0 is too big to fit 512k)\fP. 223 If using \fB\-partition_limit\fP is not enough to meet the 512k constraint, one 224 should use less segments in order to save more header bits per macroblock. 225 See the \fB\-segments\fP option. Note the \fB-m\fP and \fB-q\fP options also 226 influence the encoder's decisions and ability to hit this limit. 227 228 .SS LOGGING OPTIONS 229 These options control the level of output: 230 .TP 231 .B \-v 232 Print extra information (encoding time in particular). 233 .TP 234 .B \-print_psnr 235 Compute and report average PSNR (Peak\-Signal\-To\-Noise ratio). 236 .TP 237 .B \-print_ssim 238 Compute and report average SSIM (structural similarity 239 metric, see https://en.wikipedia.org/wiki/SSIM for additional details). 240 .TP 241 .B \-print_lsim 242 Compute and report local similarity metric (sum of lowest error amongst the 243 collocated pixel neighbors). 244 .TP 245 .B \-progress 246 Report encoding progress in percent. 247 .TP 248 .B \-quiet 249 Do not print anything. 250 .TP 251 .B \-short 252 Only print brief information (output file size and PSNR) for testing purposes. 253 .TP 254 .BI \-map " int 255 Output additional ASCII\-map of encoding information. Possible map values 256 range from 1 to 6. This is only meant to help debugging. 257 258 .SS ADDITIONAL OPTIONS 259 More advanced options are: 260 .TP 261 .BI \-s " width height 262 Specify that the input file actually consists of raw Y'CbCr samples following 263 the ITU\-R BT.601 recommendation, in 4:2:0 linear format. 264 The luma plane has size \fBwidth\fP x \fBheight\fP. 265 .TP 266 .BI \-pre " int 267 Specify some preprocessing steps. Using a value of '2' will trigger 268 quality\-dependent pseudo\-random dithering during RGBA\->YUVA conversion 269 (lossy compression only). 270 .TP 271 .BI \-alpha_filter " string 272 Specify the predictive filtering method for the alpha plane. One of 'none', 273 \&'fast' or 'best', in increasing complexity and slowness order. Default is 274 \&'fast'. Internally, alpha filtering is performed using four possible 275 predictions (none, horizontal, vertical, gradient). The 'best' mode will try 276 each mode in turn and pick the one which gives the smaller size. The 'fast' 277 mode will just try to form an a priori guess without testing all modes. 278 .TP 279 .BI \-alpha_method " int 280 Specify the algorithm used for alpha compression: 0 or 1. Algorithm 0 denotes 281 no compression, 1 uses WebP lossless format for compression. The default is 1. 282 .TP 283 .B \-exact 284 Preserve RGB values in transparent area. The default is off, to help 285 compressibility. 286 .TP 287 .BI \-blend_alpha " int 288 This option blends the alpha channel (if present) with the source using the 289 background color specified in hexadecimal as 0xrrggbb. The alpha channel is 290 afterward reset to the opaque value 255. 291 .TP 292 .B \-noalpha 293 Using this option will discard the alpha channel. 294 .TP 295 .BI \-hint " string 296 Specify the hint about input image type. Possible values are: 297 \fBphoto\fP, \fBpicture\fP or \fBgraph\fP. 298 .TP 299 .BI \-metadata " string 300 A comma separated list of metadata to copy from the input to the output if 301 present. 302 Valid values: \fBall\fP, \fBnone\fP, \fBexif\fP, \fBicc\fP, \fBxmp\fP. 303 The default is \fBnone\fP. 304 305 Note: each input format may not support all combinations. 306 .TP 307 .B \-noasm 308 Disable all assembly optimizations. 309 310 .SH EXIT STATUS 311 If there were no problems during execution, \fBcwebp\fP exits with the value of 312 the C constant \fBEXIT_SUCCESS\fP. This is usually zero. 313 .PP 314 If an error occurs, \fBcwebp\fP exits with the value of the C constant 315 \fBEXIT_FAILURE\fP. This is usually one. 316 317 .SH EXAMPLES 318 cwebp \-q 50 -lossless picture.png \-o picture_lossless.webp 319 .br 320 cwebp \-q 70 picture_with_alpha.png \-o picture_with_alpha.webp 321 .br 322 cwebp \-sns 70 \-f 50 \-size 60000 picture.png \-o picture.webp 323 .br 324 cwebp \-o picture.webp \-\- \-\-\-picture.png 325 326 .SH AUTHORS 327 \fBcwebp\fP is a part of libwebp and was written by the WebP team. 328 .br 329 The latest source tree is available at 330 https://chromium.googlesource.com/webm/libwebp 331 .PP 332 This manual page was written by Pascal Massimino <pascal.massimino@gmail.com>, 333 for the Debian project (and may be used by others). 334 335 .SH REPORTING BUGS 336 Please report all bugs to the issue tracker: 337 https://issues.webmproject.org 338 .br 339 Patches welcome! See this page to get started: 340 https://www.webmproject.org/code/contribute/submitting\-patches/ 341 342 .SH SEE ALSO 343 .BR dwebp (1), 344 .BR gif2webp (1) 345 .br 346 Please refer to https://developers.google.com/speed/webp/ for additional 347 information.