280 lines · plain
1.. SPDX-License-Identifier: GFDL-1.1-no-invariants-or-later2 3******************4Compressed Formats5******************6 7 8.. _compressed-formats:9 10.. raw:: latex11 12 \small13 14.. tabularcolumns:: |p{5.8cm}|p{1.2cm}|p{10.3cm}|15 16.. cssclass:: longtable17 18.. flat-table:: Compressed Image Formats19 :header-rows: 120 :stub-columns: 021 :widths: 3 1 422 23 * - Identifier24 - Code25 - Details26 * .. _V4L2-PIX-FMT-JPEG:27 28 - ``V4L2_PIX_FMT_JPEG``29 - 'JPEG'30 - TBD. See also :ref:`VIDIOC_G_JPEGCOMP <VIDIOC_G_JPEGCOMP>`,31 :ref:`VIDIOC_S_JPEGCOMP <VIDIOC_G_JPEGCOMP>`.32 * .. _V4L2-PIX-FMT-MPEG:33 34 - ``V4L2_PIX_FMT_MPEG``35 - 'MPEG'36 - MPEG multiplexed stream. The actual format is determined by37 extended control ``V4L2_CID_MPEG_STREAM_TYPE``, see38 :ref:`mpeg-control-id`.39 * .. _V4L2-PIX-FMT-H264:40 41 - ``V4L2_PIX_FMT_H264``42 - 'H264'43 - H264 Access Unit.44 The decoder expects one Access Unit per buffer.45 The encoder generates one Access Unit per buffer.46 If :ref:`VIDIOC_ENUM_FMT` reports ``V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM``47 then the decoder has no requirements since it can parse all the48 information from the raw bytestream.49 * .. _V4L2-PIX-FMT-H264-NO-SC:50 51 - ``V4L2_PIX_FMT_H264_NO_SC``52 - 'AVC1'53 - H264 video elementary stream without start codes.54 * .. _V4L2-PIX-FMT-H264-MVC:55 56 - ``V4L2_PIX_FMT_H264_MVC``57 - 'M264'58 - H264 MVC video elementary stream.59 * .. _V4L2-PIX-FMT-H264-SLICE:60 61 - ``V4L2_PIX_FMT_H264_SLICE``62 - 'S264'63 - H264 parsed slice data, including slice headers, either with or64 without the start code, as extracted from the H264 bitstream.65 This format is adapted for stateless video decoders that implement an66 H264 pipeline with the :ref:`stateless_decoder`.67 This pixelformat has two modifiers that must be set at least once68 through the ``V4L2_CID_STATELESS_H264_DECODE_MODE``69 and ``V4L2_CID_STATELESS_H264_START_CODE`` controls.70 In addition, metadata associated with the frame to decode are71 required to be passed through the ``V4L2_CID_STATELESS_H264_SPS``,72 ``V4L2_CID_STATELESS_H264_PPS``,73 ``V4L2_CID_STATELESS_H264_SCALING_MATRIX``,74 ``V4L2_CID_STATELESS_H264_SLICE_PARAMS`` and75 ``V4L2_CID_STATELESS_H264_DECODE_PARAMS`` controls. See the76 :ref:`associated Codec Control IDs <v4l2-codec-stateless-h264>`.77 Exactly one output and one capture buffer must be provided for use78 with this pixel format. The output buffer must contain the79 appropriate number of macroblocks to decode a full80 corresponding frame to the matching capture buffer.81 82 The syntax for this format is documented in :ref:`h264`, section83 7.3.2.8 "Slice layer without partitioning RBSP syntax" and the following84 sections.85 86 * .. _V4L2-PIX-FMT-H263:87 88 - ``V4L2_PIX_FMT_H263``89 - 'H263'90 - H263 video elementary stream.91 * .. _V4L2-PIX-FMT-SPK:92 93 - ``V4L2_PIX_FMT_SPK``94 - 'SPK0'95 - Sorenson Spark is an implementation of H.263 for use in Flash Video and Adobe Flash files96 * .. _V4L2-PIX-FMT-MPEG1:97 98 - ``V4L2_PIX_FMT_MPEG1``99 - 'MPG1'100 - MPEG1 Picture. Each buffer starts with a Picture header, followed101 by other headers as needed and ending with the Picture data.102 If :ref:`VIDIOC_ENUM_FMT` reports ``V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM``103 then the decoder has no requirements since it can parse all the104 information from the raw bytestream.105 * .. _V4L2-PIX-FMT-MPEG2:106 107 - ``V4L2_PIX_FMT_MPEG2``108 - 'MPG2'109 - MPEG2 Picture. Each buffer starts with a Picture header, followed110 by other headers as needed and ending with the Picture data.111 If :ref:`VIDIOC_ENUM_FMT` reports ``V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM``112 then the decoder has no requirements since it can parse all the113 information from the raw bytestream.114 * .. _V4L2-PIX-FMT-MPEG2-SLICE:115 116 - ``V4L2_PIX_FMT_MPEG2_SLICE``117 - 'MG2S'118 - MPEG-2 parsed slice data, as extracted from the MPEG-2 bitstream.119 This format is adapted for stateless video decoders that implement a120 MPEG-2 pipeline with the :ref:`stateless_decoder`.121 Metadata associated with the frame to decode is required to be passed122 through the ``V4L2_CID_STATELESS_MPEG2_SEQUENCE`` and123 ``V4L2_CID_STATELESS_MPEG2_PICTURE`` controls.124 Quantisation matrices can optionally be specified through the125 ``V4L2_CID_STATELESS_MPEG2_QUANTISATION`` control.126 See the :ref:`associated Codec Control IDs <v4l2-codec-stateless-mpeg2>`.127 Exactly one output and one capture buffer must be provided for use with128 this pixel format. The output buffer must contain the appropriate number129 of macroblocks to decode a full corresponding frame to the matching130 capture buffer.131 * .. _V4L2-PIX-FMT-MPEG4:132 133 - ``V4L2_PIX_FMT_MPEG4``134 - 'MPG4'135 - MPEG4 video elementary stream.136 * .. _V4L2-PIX-FMT-XVID:137 138 - ``V4L2_PIX_FMT_XVID``139 - 'XVID'140 - Xvid video elementary stream.141 * .. _V4L2-PIX-FMT-VC1-ANNEX-G:142 143 - ``V4L2_PIX_FMT_VC1_ANNEX_G``144 - 'VC1G'145 - VC1, SMPTE 421M Annex G compliant stream.146 * .. _V4L2-PIX-FMT-VC1-ANNEX-L:147 148 - ``V4L2_PIX_FMT_VC1_ANNEX_L``149 - 'VC1L'150 - VC1, SMPTE 421M Annex L compliant stream.151 * .. _V4L2-PIX-FMT-VP8:152 153 - ``V4L2_PIX_FMT_VP8``154 - 'VP80'155 - VP8 compressed video frame. The encoder generates one156 compressed frame per buffer, and the decoder requires one157 compressed frame per buffer.158 * .. _V4L2-PIX-FMT-VP8-FRAME:159 160 - ``V4L2_PIX_FMT_VP8_FRAME``161 - 'VP8F'162 - VP8 parsed frame, including the frame header, as extracted from the container.163 This format is adapted for stateless video decoders that implement an164 VP8 pipeline with the :ref:`stateless_decoder`.165 Metadata associated with the frame to decode is required to be passed166 through the ``V4L2_CID_STATELESS_VP8_FRAME`` control.167 See the :ref:`associated Codec Control IDs <v4l2-codec-stateless-vp8>`.168 Exactly one output and one capture buffer must be provided for use with169 this pixel format. The output buffer must contain the appropriate number170 of macroblocks to decode a full corresponding frame to the matching171 capture buffer.172 173 * .. _V4L2-PIX-FMT-VP9:174 175 - ``V4L2_PIX_FMT_VP9``176 - 'VP90'177 - VP9 compressed video frame. The encoder generates one178 compressed frame per buffer, and the decoder requires one179 compressed frame per buffer.180 * .. _V4L2-PIX-FMT-VP9-FRAME:181 182 - ``V4L2_PIX_FMT_VP9_FRAME``183 - 'VP9F'184 - VP9 parsed frame, including the frame header, as extracted from the container.185 This format is adapted for stateless video decoders that implement a186 VP9 pipeline with the :ref:`stateless_decoder`.187 Metadata associated with the frame to decode is required to be passed188 through the ``V4L2_CID_STATELESS_VP9_FRAME`` and189 the ``V4L2_CID_STATELESS_VP9_COMPRESSED_HDR`` controls.190 See the :ref:`associated Codec Control IDs <v4l2-codec-stateless-vp9>`.191 Exactly one output and one capture buffer must be provided for use with192 this pixel format. The output buffer must contain the appropriate number193 of macroblocks to decode a full corresponding frame to the matching194 capture buffer.195 * .. _V4L2-PIX-FMT-HEVC:196 197 - ``V4L2_PIX_FMT_HEVC``198 - 'HEVC'199 - HEVC/H.265 Access Unit.200 The decoder expects one Access Unit per buffer.201 The encoder generates one Access Unit per buffer.202 If :ref:`VIDIOC_ENUM_FMT` reports ``V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM``203 then the decoder has no requirements since it can parse all the204 information from the raw bytestream.205 * .. _V4L2-PIX-FMT-HEVC-SLICE:206 207 - ``V4L2_PIX_FMT_HEVC_SLICE``208 - 'S265'209 - HEVC parsed slice data, as extracted from the HEVC bitstream.210 This format is adapted for stateless video decoders that implement a211 HEVC pipeline (using the :ref:`mem2mem` and :ref:`media-request-api`).212 This pixelformat has two modifiers that must be set at least once213 through the ``V4L2_CID_MPEG_VIDEO_HEVC_DECODE_MODE``214 and ``V4L2_CID_MPEG_VIDEO_HEVC_START_CODE`` controls.215 Metadata associated with the frame to decode is required to be passed216 through the following controls:217 ``V4L2_CID_MPEG_VIDEO_HEVC_SPS``,218 ``V4L2_CID_MPEG_VIDEO_HEVC_PPS``, and219 ``V4L2_CID_MPEG_VIDEO_HEVC_SLICE_PARAMS``.220 See the :ref:`associated Codec Control IDs <v4l2-codec-stateless-hevc>`.221 Buffers associated with this pixel format must contain the appropriate222 number of macroblocks to decode a full corresponding frame.223 * .. _V4L2-PIX-FMT-FWHT:224 225 - ``V4L2_PIX_FMT_FWHT``226 - 'FWHT'227 - Video elementary stream using a codec based on the Fast Walsh Hadamard228 Transform. This codec is implemented by the vicodec ('Virtual Codec')229 driver. See the codec-fwht.h header for more details.230 :ref:`VIDIOC_ENUM_FMT` reports ``V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM``231 since the decoder can parse all the information from the raw bytestream.232 * .. _V4L2-PIX-FMT-FWHT-STATELESS:233 234 - ``V4L2_PIX_FMT_FWHT_STATELESS``235 - 'SFWH'236 - Same format as V4L2_PIX_FMT_FWHT but requires stateless codec implementation.237 Metadata associated with the frame to decode is required to be passed238 through the ``V4L2_CID_STATELESS_FWHT_PARAMS`` control.239 See the :ref:`associated Codec Control ID <codec-stateless-fwht>`.240 * .. _V4L2-PIX-FMT-RV30:241 242 - ``V4L2_PIX_FMT_RV30``243 - 'RV30'244 - RealVideo, or also spelled as Real Video, is a suite of245 proprietary video compression formats developed by246 RealNetworks - the specific format changes with the version.247 RealVideo codecs are identified by four-character codes.248 RV30 corresponds to RealVideo 8, suspected to be based249 largely on an early draft of H.264250 * .. _V4L2-PIX-FMT-RV40:251 252 - ``V4L2_PIX_FMT_RV40``253 - 'RV40'254 - RV40 represents RealVideo 9 and RealVideo 10.255 RealVideo 9, suspected to be based on H.264.256 RealVideo 10, aka RV9 EHQ, This refers to an improved encoder257 for the RV9 format that is fully backwards compatible with258 RV9 players - the format and decoder did not change, only259 the encoder did. As a result, it uses the same FourCC.260 261 * .. _V4L2-PIX-FMT-AV1-FRAME:262 263 - ``V4L2_PIX_FMT_AV1_FRAME``264 - 'AV1F'265 - AV1 parsed frame, including the frame header, as extracted from the container.266 This format is adapted for stateless video decoders that implement a AV1267 pipeline with the :ref:`stateless_decoder`. Metadata associated with the268 frame to decode is required to be passed through the269 ``V4L2_CID_STATELESS_AV1_SEQUENCE``, ``V4L2_CID_STATELESS_AV1_FRAME``,270 and ``V4L2_CID_STATELESS_AV1_TILE_GROUP_ENTRY`` controls.271 See the :ref:`associated Codec Control IDs <v4l2-codec-stateless-av1>`.272 Exactly one output and one capture buffer must be provided for use with273 this pixel format. The output buffer must contain the appropriate number274 of macroblocks to decode a full corresponding frame to the matching275 capture buffer.276 277.. raw:: latex278 279 \normalsize280