文档章节

FFmpeg结构体彻底分析——AVFrame

zhangyujsj
 zhangyujsj
发布于 2014/12/17 16:34
字数 1665
阅读 222
收藏 0

/**
 * This structure describes decoded (raw) audio or video data.
 *
 * AVFrame must be allocated using av_frame_alloc(). Note that this only
 * allocates the AVFrame itself, the buffers for the data must be managed
 * through other means (see below).
 * AVFrame must be freed with av_frame_free().

AVFrame结构体描述原始视频数据。

分配方法:av_frame_alloc(),特别注意这个方法仅仅分配了AVFrame自己其中用到的缓冲区是不去分配的

释放方法:av_frame_free()

AVFrame是典型的一次分配反复使用来装不同的数据,在这种情况下av_frame_unref()函数将在重新使用之前释放任何由frame持有的引用并且再次重置到它的原始干净状态
 *
 * AVFrame is typically allocated once and then reused multiple times to hold
 * different data (e.g. a single AVFrame to hold frames received from a
 * decoder). In such a case, av_frame_unref() will free any references held by
 * the frame and reset it to its original clean state before it
 * is reused again.
 *
 * The data described by an AVFrame is usually reference counted through the
 * AVBuffer API. The underlying buffer references are stored in AVFrame.buf /
 * AVFrame.extended_buf. An AVFrame is considered to be reference counted if at
 * least one reference is set, i.e. if AVFrame.buf[0] != NULL. In such a case,
 * every single data plane must be contained in one of the buffers in
 * AVFrame.buf or AVFrame.extended_buf.
 * There may be a single buffer for all the data, or one separate buffer for
 * each plane, or anything in between.
 *
 * sizeof(AVFrame) is not a part of the public ABI, so new fields may be added
 * to the end with a minor bump.
 * Similarly fields that are marked as to be only accessed by
 * av_opt_ptr() can be reordered. This allows 2 forks to add fields
 * without breaking compatibility with each other.
 */
typedef struct AVFrame {
#define AV_NUM_DATA_POINTERS 8
    /**
     * pointer to the picture/channel planes.
     * This might be different from the first allocated byte
     *
     * Some decoders access areas outside 0,0 - width,height, please
     * see avcodec_align_dimensions2(). Some filters and swscale can read
     * up to 16 bytes beyond the planes, if these filters are to be used,
     * then 16 extra bytes must be allocated.
     */
    uint8_t *data[AV_NUM_DATA_POINTERS];//到图像或者声道平面的指针

    /**
     * For video, size in bytes of each picture line.
     * For audio, size in bytes of each plane.
     *
     * For audio, only linesize[0] may be set. For planar audio, each channel
     * plane must be the same size.
     *
     * For video the linesizes should be multiplies of the CPUs alignment
     * preference, this is 16 or 32 for modern desktop CPUs.
     * Some code requires such alignment other code can be slower without
     * correct alignment, for yet other it makes no difference.
     *
     * @note The linesize may be larger than the size of usable data -- there
     * may be extra padding present for performance reasons.
     */
    int linesize[AV_NUM_DATA_POINTERS];//对视频:每一个图片行的大小字节单位;对音频:每一个plane的字节单位的大小

    /**
     * pointers to the data planes/channels.
     *
     * For video, this should simply point to data[].
     *
     * For planar audio, each channel has a separate data pointer, and
     * linesize[0] contains the size of each channel buffer.
     * For packed audio, there is just one data pointer, and linesize[0]
     * contains the total size of the buffer for all channels.
     *
     * Note: Both data and extended_data should always be set in a valid frame,
     * but for planar audio with more channels that can fit in data,
     * extended_data must be used in order to access all channels.
     */
    uint8_t **extended_data;

    /**
     * width and height of the video frame
     */
    int width, height;//视频帧的宽高

    /**
     * number of audio samples (per channel) described by this frame
     */
    int nb_samples;

    /**
     * format of the frame, -1 if unknown or unset
     * Values correspond to enum AVPixelFormat for video frames,
     * enum AVSampleFormat for audio)
     */
    int format;//帧的格式AVPixelFormat或者AVSampleFormat

    /**
     * 1 -> keyframe, 0-> not
     */
    int key_frame;//关键帧

    /**
     * Picture type of the frame.
     */
    enum AVPictureType pict_type;//帧的图像类型

#if FF_API_AVFRAME_LAVC
    attribute_deprecated
    uint8_t *base[AV_NUM_DATA_POINTERS];
#endif

    /**
     * Sample aspect ratio for the video frame, 0/1 if unknown/unspecified.
     */
    AVRational sample_aspect_ratio;

    /**
     * Presentation timestamp in time_base units (time when frame should be shown to user).
     */
    int64_t pts;//显示的时间戳,就是指出这个帧什么时候显示给用户

    /**
     * PTS copied from the AVPacket that was decoded to produce this frame.
     */
    int64_t pkt_pts;//从解码产生这个AVFrame的AVPacket那里copy过来的pts

    /**
     * DTS copied from the AVPacket that triggered returning this frame. (if frame threading isnt used)
     * This is also the Presentation time of this AVFrame calculated from
     * only AVPacket.dts values without pts values.
     */
    int64_t pkt_dts;

    /**
     * picture number in bitstream order
     */
    int coded_picture_number;
    /**
     * picture number in display order
     */
    int display_picture_number;

    /**
     * quality (between 1 (good) and FF_LAMBDA_MAX (bad))
     */
    int quality;

#if FF_API_AVFRAME_LAVC
    attribute_deprecated
    int reference;

    /**
     * QP table
     */
    attribute_deprecated
    int8_t *qscale_table;
    /**
     * QP store stride
     */
    attribute_deprecated
    int qstride;

    attribute_deprecated
    int qscale_type;

    /**
     * mbskip_table[mb]>=1 if MB didn't change
     * stride= mb_width = (width+15)>>4
     */
    attribute_deprecated
    uint8_t *mbskip_table;

    /**
     * motion vector table
     * @code
     * example:
     * int mv_sample_log2= 4 - motion_subsample_log2;
     * int mb_width= (width+15)>>4;
     * int mv_stride= (mb_width << mv_sample_log2) + 1;
     * motion_val[direction][x + y*mv_stride][0->mv_x, 1->mv_y];
     * @endcode
     */
    int16_t (*motion_val[2])[2];

    /**
     * macroblock type table
     * mb_type_base + mb_width + 2
     */
    attribute_deprecated
    uint32_t *mb_type;

    /**
     * DCT coefficients
     */
    attribute_deprecated
    short *dct_coeff;

    /**
     * motion reference frame index
     * the order in which these are stored can depend on the codec.
     */
    attribute_deprecated
    int8_t *ref_index[2];
#endif

    /**
     * for some private data of the user
     */
    void *opaque;

    /**
     * error
     */
    uint64_t error[AV_NUM_DATA_POINTERS];

#if FF_API_AVFRAME_LAVC
    attribute_deprecated
    int type;
#endif

    /**
     * When decoding, this signals how much the picture must be delayed.
     * extra_delay = repeat_pict / (2*fps)
     */
    int repeat_pict;

    /**
     * The content of the picture is interlaced.
     */
    int interlaced_frame;

    /**
     * If the content is interlaced, is top field displayed first.
     */
    int top_field_first;

    /**
     * Tell user application that palette has changed from previous frame.
     */
    int palette_has_changed;

#if FF_API_AVFRAME_LAVC
    attribute_deprecated
    int buffer_hints;

    /**
     * Pan scan.
     */
    attribute_deprecated
    struct AVPanScan *pan_scan;
#endif

    /**
     * reordered opaque 64bit (generally an integer or a double precision float
     * PTS but can be anything).
     * The user sets AVCodecContext.reordered_opaque to represent the input at
     * that time,
     * the decoder reorders values as needed and sets AVFrame.reordered_opaque
     * to exactly one of the values provided by the user through AVCodecContext.reordered_opaque
     * @deprecated in favor of pkt_pts
     */
    int64_t reordered_opaque;

#if FF_API_AVFRAME_LAVC
    /**
     * @deprecated this field is unused
     */
    attribute_deprecated void *hwaccel_picture_private;

    attribute_deprecated
    struct AVCodecContext *owner;
    attribute_deprecated
    void *thread_opaque;

    /**
     * log2 of the size of the block which a single vector in motion_val represents:
     * (4->16x16, 3->8x8, 2-> 4x4, 1-> 2x2)
     */
    uint8_t motion_subsample_log2;
#endif

    /**
     * Sample rate of the audio data.
     */
    int sample_rate;

    /**
     * Channel layout of the audio data.
     */
    uint64_t channel_layout;

    /**
     * AVBuffer references backing the data for this frame. If all elements of
     * this array are NULL, then this frame is not reference counted.
     *
     * There may be at most one AVBuffer per data plane, so for video this array
     * always contains all the references. For planar audio with more than
     * AV_NUM_DATA_POINTERS channels, there may be more buffers than can fit in
     * this array. Then the extra AVBufferRef pointers are stored in the
     * extended_buf array.
     */
    AVBufferRef *buf[AV_NUM_DATA_POINTERS];

    /**
     * For planar audio which requires more than AV_NUM_DATA_POINTERS
     * AVBufferRef pointers, this array will hold all the references which
     * cannot fit into AVFrame.buf.
     *
     * Note that this is different from AVFrame.extended_data, which always
     * contains all the pointers. This array only contains the extra pointers,
     * which cannot fit into AVFrame.buf.
     *
     * This array is always allocated using av_malloc() by whoever constructs
     * the frame. It is freed in av_frame_unref().
     */
    AVBufferRef **extended_buf;
    /**
     * Number of elements in extended_buf.
     */
    int        nb_extended_buf;

    AVFrameSideData **side_data;
    int            nb_side_data;

/**
 * @defgroup lavu_frame_flags AV_FRAME_FLAGS
 * Flags describing additional frame properties.
 *
 * @{
 */

/**
 * The frame data may be corrupted, e.g. due to decoding errors.
 */
#define AV_FRAME_FLAG_CORRUPT       (1 << 0)
/**
 * @}
 */

    /**
     * Frame flags, a combination of @ref lavu_frame_flags
     */
    int flags;

    /**
     * frame timestamp estimated using various heuristics, in stream time base
     * Code outside libavcodec should access this field using:
     * av_frame_get_best_effort_timestamp(frame)
     * - encoding: unused
     * - decoding: set by libavcodec, read by user.
     */
    int64_t best_effort_timestamp;

    /**
     * reordered pos from the last AVPacket that has been input into the decoder
     * Code outside libavcodec should access this field using:
     * av_frame_get_pkt_pos(frame)
     * - encoding: unused
     * - decoding: Read by user.
     */
    int64_t pkt_pos;

    /**
     * duration of the corresponding packet, expressed in
     * AVStream->time_base units, 0 if unknown.
     * Code outside libavcodec should access this field using:
     * av_frame_get_pkt_duration(frame)
     * - encoding: unused
     * - decoding: Read by user.
     */
    int64_t pkt_duration;

    /**
     * metadata.
     * Code outside libavcodec should access this field using:
     * av_frame_get_metadata(frame)
     * - encoding: Set by user.
     * - decoding: Set by libavcodec.
     */
    AVDictionary *metadata;

    /**
     * decode error flags of the frame, set to a combination of
     * FF_DECODE_ERROR_xxx flags if the decoder produced a frame, but there
     * were errors during the decoding.
     * Code outside libavcodec should access this field using:
     * av_frame_get_decode_error_flags(frame)
     * - encoding: unused
     * - decoding: set by libavcodec, read by user.
     */
    int decode_error_flags;
#define FF_DECODE_ERROR_INVALID_BITSTREAM   1
#define FF_DECODE_ERROR_MISSING_REFERENCE   2

    /**
     * number of audio channels, only used for audio.
     * Code outside libavcodec should access this field using:
     * av_frame_get_channels(frame)
     * - encoding: unused
     * - decoding: Read by user.
     */
    int channels;

    /**
     * size of the corresponding packet containing the compressed
     * frame. It must be accessed using av_frame_get_pkt_size() and
     * av_frame_set_pkt_size().
     * It is set to a negative value if unknown.
     * - encoding: unused
     * - decoding: set by libavcodec, read by user.
     */
    int pkt_size;

    /**
     * YUV colorspace type.
     * It must be accessed using av_frame_get_colorspace() and
     * av_frame_set_colorspace().
     * - encoding: Set by user
     * - decoding: Set by libavcodec
     */
    enum AVColorSpace colorspace;

    /**
     * MPEG vs JPEG YUV range.
     * It must be accessed using av_frame_get_color_range() and
     * av_frame_set_color_range().
     * - encoding: Set by user
     * - decoding: Set by libavcodec
     */
    enum AVColorRange color_range;


    /**
     * Not to be accessed directly from outside libavutil
     */
    AVBufferRef *qp_table_buf;
} AVFrame;

© 著作权归作者所有

zhangyujsj
粉丝 24
博文 358
码字总数 224241
作品 0
广州
私信 提问
使用 liavformat 和 libavcodec 实现解码器

使用ffmpeg 的liavformat 解封装,使用libavcodec 解codec,实现一个简单的解码器。 解码的流程,从数据结构上看就是AVFormatContext ->AVCodecContext -> AVPacket -> AVFrame。 生成AVPac...

鼎铭
2018/07/09
0
0
FFmpeg命令行工具学习(五):FFmpeg 编解码 API 分析

在上一篇文章 FFmpeg命令行工具学习(四):FFmpeg API 介绍与通用 API 分析 中,我们简单的讲解了一下FFmpeg 的API基本概念,并分析了一下通用API,本文我们将分析 FFmpeg 在编解码时使用的A...

灰色飘零
2018/07/20
0
0
FFmpeg命令行工具学习(四):FFmpeg API 介绍与通用 API 分析

一、FFmpeg 相关术语 1. 容器/文件(Container/File):即特定格式的多媒体文件,比如MP4,flv,mov等。 2. 媒体流(Stream):表示在时间轴上的一段连续的数据,比如一段声音数据、一段视频...

灰色飘零
2018/07/17
0
0
最新FFmpeg音视频核心技术精讲与实战(完整版)

第1章 课程导学与准备工作 全民娱乐时代,需要音视频等多媒体产品层出不穷,但会处理音视频数据的工程师却极度匮乏,进入音视频开发领域正当时,这门课程就是为这样的你而生。来吧!加入我们...

L272339
01/06
0
0
2018FFmpeg音视频核心技术精讲与实战(已完结)目前最新

第1章 课程导学与准备工作 全民娱乐时代,需要音视频等多媒体产品层出不穷,但会处理音视频数据的工程师却极度匮乏,进入音视频开发领域正当时,这门课程就是为这样的你而生。来吧!加入我们...

聂友华
2018/12/18
0
0

没有更多内容

加载失败,请刷新页面

加载更多

EasyExcel

<!-- alibaba EasyExcel escel导入导出 --> <dependency> <groupId>com.alibaba</groupId> <artifactId>easyexcel</artifactId> <version>1.0.0-RELEASE</version> </dependency>......

少年已不再年少
20分钟前
0
0
解决:warning: deprecated conversion from string constant to 'char*' [-Wwrite-strings]

代码 static int CheckOneReg(uint8_t busNo, char *title, int32_t addr, uint32_t *val, BOOL isPrintfNL, ErrMsg_t const *pMsg)res |= CheckOneReg(busNo,"fwVer:", HALREG_FW_VER, ......

SamXIAO
20分钟前
0
0
jenkins搭建全流程

安装环境   操作系统:linux   软件:jdk 8   软件:tomcat(apache-tomcat-7.0.90) 软件:jenkins 软件:maven 软件:git 1.1安装步骤 1.1.1安装jdk 由于Jenkins是基于Java开发的,安...

shzwork
26分钟前
0
0
Java的战争

本文来自微信公众号: 金捷幡(ID:jin-jiefan) ,作者:金捷幡,封面:拉里·埃里森(东方IC) 2019年5月,彻底撕破脸的特朗普掀起对华为的战争,谷歌被迫吊销了华为的Android授权。开源软...

Java领航员
今天
8
0
超详细的LM3414MRX/NOPB规格参数介绍就在这里

超详细的LM3414MRX/NOPB规格参数介绍就在这里 描述 LM3414和LM3414MRX/NOPB是具有1-A 60 W(1)共阳极功能的恒流降压LED驱动器。它们适用于驱动单串3-W HBLED,效率高达96%。它们可接受4.5...

不能吃肉的仙女
今天
3
0

没有更多内容

加载失败,请刷新页面

加载更多

返回顶部
顶部