分享一下我老师大神的人工智能教程!零基础,通俗易懂!http://blog.csdn.net/jiangjunshow
也欢迎大家转载本篇文章。分享知识,造福人民,实现我们中华民族伟大复兴!
=====================================================
FFmpeg的库函数源代码分析文章列表:
【架构图】
【通用】
FFmpeg 源代码简单分析:av_register_all()
FFmpeg 源代码简单分析:avcodec_register_all()
FFmpeg 源代码简单分析:内存的分配和释放(av_malloc()、av_free()等)
FFmpeg 源代码简单分析:常见结构体的初始化和销毁(AVFormatContext,AVFrame等)
FFmpeg 源代码简单分析:av_find_decoder()和av_find_encoder()
FFmpeg 源代码简单分析:avcodec_open2()
FFmpeg 源代码简单分析:avcodec_close()
【解码】
图解FFMPEG打开媒体的函数avformat_open_input
FFmpeg 源代码简单分析:avformat_open_input()
FFmpeg 源代码简单分析:avformat_find_stream_info()
FFmpeg 源代码简单分析:av_read_frame()
FFmpeg 源代码简单分析:avcodec_decode_video2()
FFmpeg 源代码简单分析:avformat_close_input()
【编码】
FFmpeg 源代码简单分析:avformat_alloc_output_context2()
FFmpeg 源代码简单分析:avformat_write_header()
FFmpeg 源代码简单分析:avcodec_encode_video()
FFmpeg 源代码简单分析:av_write_frame()
FFmpeg 源代码简单分析:av_write_trailer()
【其它】
FFmpeg源代码简单分析:日志输出系统(av_log()等)
FFmpeg源代码简单分析:结构体成员管理系统-AVClass
FFmpeg源代码简单分析:结构体成员管理系统-AVOption
FFmpeg源代码简单分析:libswscale的sws_getContext()
FFmpeg源代码简单分析:libswscale的sws_scale()
FFmpeg源代码简单分析:libavdevice的avdevice_register_all()
FFmpeg源代码简单分析:libavdevice的gdigrab
【脚本】
【H.264】
=====================================================
ffmpeg中的avcodec_decode_video2()的作用是解码一帧视频数据。输入一个压缩编码的结构体AVPacket,输出一个解码后的结构体AVFrame。该函数的声明位于libavcodec\avcodec.h,如下所示。
/** * Decode the video frame of size avpkt->size from avpkt->data into picture. * Some decoders may support multiple frames in a single AVPacket, such * decoders would then just decode the first frame. * * @warning The input buffer must be FF_INPUT_BUFFER_PADDING_SIZE larger than * the actual read bytes because some optimized bitstream readers read 32 or 64 * bits at once and could read over the end. * * @warning The end of the input buffer buf should be set to 0 to ensure that * no overreading happens for damaged MPEG streams. * * @note Codecs which have the CODEC_CAP_DELAY capability set have a delay * between input and output, these need to be fed with avpkt->data=NULL, * avpkt->size=0 at the end to return the remaining frames. * * @param avctx the codec context * @param[out] picture The AVFrame in which the decoded video frame will be stored. * Use av_frame_alloc() to get an AVFrame. The codec will * allocate memory for the actual bitmap by calling the * AVCodecContext.get_buffer2() callback. * When AVCodecContext.refcounted_frames is set to 1, the frame is * reference counted and the returned reference belongs to the * caller. The caller must release the frame using av_frame_unref() * when the frame is no longer needed. The caller may safely write * to the frame if av_frame_is_writable() returns 1. * When AVCodecContext.refcounted_frames is set to 0, the returned * reference belongs to the decoder and is valid only until the * next call to this function or until closing or flushing the * decoder. The caller may not write to it. * * @param[in] avpkt The input AVPacket containing the input buffer. * You can create such packet with av_init_packet() and by then setting * data and size, some decoders might in addition need other fields like * flags&AV_PKT_FLAG_KEY. All decoders are designed to use the least * fields possible. * @param[in,out] got_picture_ptr Zero if no frame could be decompressed, otherwise, it is nonzero. * @return On error a negative value is returned, otherwise the number of bytes * used or zero if no frame could be decompressed. */int avcodec_decode_video2(AVCodecContext *avctx, AVFrame *picture, int *got_picture_ptr, const AVPacket *avpkt);
查看源代码之后发现,这个函数竟然十分的简单,源代码位于libavcodec\utils.c,如下所示:
int attribute_align_arg avcodec_decode_video2(AVCodecContext *avctx, AVFrame *picture, int *got_picture_ptr, const AVPacket *avpkt){ AVCodecInternal *avci = avctx->internal; int ret; // copy to ensure we do not change avpkt AVPacket tmp = *avpkt; if (!avctx->codec) return AVERROR(EINVAL); //检查是不是视频(非音频) if (avctx->codec->type != AVMEDIA_TYPE_VIDEO) { av_log(avctx, AV_LOG_ERROR, "Invalid media type for video\n"); return AVERROR(EINVAL); } *got_picture_ptr = 0; //检查宽、高设置是否正确 if ((avctx->coded_width || avctx->coded_height) && av_image_check_size(avctx->coded_width, avctx->coded_height, 0, avctx)) return AVERROR(EINVAL); av_frame_unref(picture); if ((avctx->codec->capabilities & CODEC_CAP_DELAY) || avpkt->size || (avctx->active_thread_type & FF_THREAD_FRAME)) { int did_split = av_packet_split_side_data(&tmp); ret = apply_param_change(avctx, &tmp); if (ret < 0) { av_log(avctx, AV_LOG_ERROR, "Error applying parameter changes.\n"); if (avctx->err_recognition & AV_EF_EXPLODE) goto fail; } avctx->internal->pkt = &tmp; if (HAVE_THREADS && avctx->active_thread_type & FF_THREAD_FRAME) ret = ff_thread_decode_frame(avctx, picture, got_picture_ptr, &tmp); else { //最关键的解码函数 ret = avctx->codec->decode(avctx, picture, got_picture_ptr, &tmp); //设置pkt_dts字段的值 picture->pkt_dts = avpkt->dts; if(!avctx->has_b_frames){ av_frame_set_pkt_pos(picture, avpkt->pos); } //FIXME these should be under if(!avctx->has_b_frames) /* get_buffer is supposed to set frame parameters */ if (!(avctx->codec->capabilities & CODEC_CAP_DR1)) { //对一些字段进行赋值 if (!picture->sample_aspect_ratio.num) picture->sample_aspect_ratio = avctx->sample_aspect_ratio; if (!picture->width) picture->width = avctx->width; if (!picture->height) picture->height = avctx->height; if (picture->format == AV_PIX_FMT_NONE) picture->format = avctx->pix_fmt; } } add_metadata_from_side_data(avctx, picture);fail: emms_c(); //needed to avoid an emms_c() call before every return; avctx->internal->pkt = NULL; if (did_split) { av_packet_free_side_data(&tmp); if(ret == tmp.size) ret = avpkt->size; } if (*got_picture_ptr) { if (!avctx->refcounted_frames) { int err = unrefcount_frame(avci, picture); if (err < 0) return err; } avctx->frame_number++; av_frame_set_best_effort_timestamp(picture, guess_correct_pts(avctx, picture->pkt_pts, picture->pkt_dts)); } else av_frame_unref(picture); } else ret = 0; /* many decoders assign whole AVFrames, thus overwriting extended_data; * make sure it's set correctly */ av_assert0(!picture->extended_data || picture->extended_data == picture->data);#if FF_API_AVCTX_TIMEBASE if (avctx->framerate.num > 0 && avctx->framerate.den > 0) avctx->time_base = av_inv_q(av_mul_q(avctx->framerate, (AVRational){avctx->ticks_per_frame, 1}));#endif return ret;}
从代码中可以看出,avcodec_decode_video2()主要做了以下几个方面的工作:
(1)对输入的字段进行了一系列的检查工作:例如宽高是否正确,输入是否为视频等等。
(2)通过ret = avctx->codec->decode(avctx, picture, got_picture_ptr,&tmp)这句代码,调用了相应AVCodec的decode()函数,完成了解码操作。
(3)对得到的AVFrame的一些字段进行了赋值,例如宽高、像素格式等等。
其中第二部是关键的一步,它调用了AVCodec的decode()方法完成了解码。AVCodec的decode()方法是一个函数指针,指向了具体解码器的解码函数。在这里我们以H.264解码器为例,看一下解码的实现过程。H.264解码器对应的AVCodec的定义位于libavcodec\h264.c,如下所示。
AVCodec ff_h264_decoder = { .name = "h264", .long_name = NULL_IF_CONFIG_SMALL("H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10"), .type = AVMEDIA_TYPE_VIDEO, .id = AV_CODEC_ID_H264, .priv_data_size = sizeof(H264Context), .init = ff_h264_decode_init, .close = h264_decode_end, .decode = h264_decode_frame, .capabilities = /*CODEC_CAP_DRAW_HORIZ_BAND |*/ CODEC_CAP_DR1 | CODEC_CAP_DELAY | CODEC_CAP_SLICE_THREADS | CODEC_CAP_FRAME_THREADS, .flush = flush_dpb, .init_thread_copy = ONLY_IF_THREADS_ENABLED(decode_init_thread_copy), .update_thread_context = ONLY_IF_THREADS_ENABLED(ff_h264_update_thread_context), .profiles = NULL_IF_CONFIG_SMALL(profiles), .priv_class = &h264_class,};
从ff_h264_decoder的定义可以看出,decode()指向了h264_decode_frame()函数。继续看一下h264_decode_frame()函数的定义,如下所示。
static int h264_decode_frame(AVCodecContext *avctx, void *data, int *got_frame, AVPacket *avpkt){ const uint8_t *buf = avpkt->data; int buf_size = avpkt->size; H264Context *h = avctx->priv_data; AVFrame *pict = data; int buf_index = 0; H264Picture *out; int i, out_idx; int ret; h->flags = avctx->flags; /* reset data partitioning here, to ensure GetBitContexts from previous * packets do not get used. */ h->data_partitioning = 0; /* end of stream, output what is still in the buffers */ if (buf_size == 0) { out: h->cur_pic_ptr = NULL; h->first_field = 0; // FIXME factorize this with the output code below out = h->delayed_pic[0]; out_idx = 0; for (i = 1; h->delayed_pic[i] && !h->delayed_pic[i]->f.key_frame && !h->delayed_pic[i]->mmco_reset; i++) if (h->delayed_pic[i]->poc < out->poc) { out = h->delayed_pic[i]; out_idx = i; } for (i = out_idx; h->delayed_pic[i]; i++) h->delayed_pic[i] = h->delayed_pic[i + 1]; if (out) { out->reference &= ~DELAYED_PIC_REF; ret = output_frame(h, pict, out); if (ret < 0) return ret; *got_frame = 1; } return buf_index; } if (h->is_avc && av_packet_get_side_data(avpkt, AV_PKT_DATA_NEW_EXTRADATA, NULL)) { int side_size; uint8_t *side = av_packet_get_side_data(avpkt, AV_PKT_DATA_NEW_EXTRADATA, &side_size); if (is_extra(side, side_size)) ff_h264_decode_extradata(h, side, side_size); } if(h->is_avc && buf_size >= 9 && buf[0]==1 && buf[2]==0 && (buf[4]&0xFC)==0xFC && (buf[5]&0x1F) && buf[8]==0x67){ if (is_extra(buf, buf_size)) return ff_h264_decode_extradata(h, buf, buf_size); } //H.264解码 buf_index = decode_nal_units(h, buf, buf_size, 0); if (buf_index < 0) return AVERROR_INVALIDDATA; if (!h->cur_pic_ptr && h->nal_unit_type == NAL_END_SEQUENCE) { av_assert0(buf_index <= buf_size); goto out; } if (!(avctx->flags2 & CODEC_FLAG2_CHUNKS) && !h->cur_pic_ptr) { if (avctx->skip_frame >= AVDISCARD_NONREF || buf_size >= 4 && !memcmp("Q264", buf, 4)) return buf_size; av_log(avctx, AV_LOG_ERROR, "no frame!\n"); return AVERROR_INVALIDDATA; } if (!(avctx->flags2 & CODEC_FLAG2_CHUNKS) || (h->mb_y >= h->mb_height && h->mb_height)) { if (avctx->flags2 & CODEC_FLAG2_CHUNKS) decode_postinit(h, 1); ff_h264_field_end(h, 0); /* Wait for second field. */ *got_frame = 0; if (h->next_output_pic && ( h->next_output_pic->recovered)) { if (!h->next_output_pic->recovered) h->next_output_pic->f.flags |= AV_FRAME_FLAG_CORRUPT; ret = output_frame(h, pict, h->next_output_pic); if (ret < 0) return ret; *got_frame = 1; if (CONFIG_MPEGVIDEO) { ff_print_debug_info2(h->avctx, pict, h->er.mbskip_table, h->next_output_pic->mb_type, h->next_output_pic->qscale_table, h->next_output_pic->motion_val, &h->low_delay, h->mb_width, h->mb_height, h->mb_stride, 1); } } } assert(pict->buf[0] || !*got_frame); return get_consumed_bytes(buf_index, buf_size);}
从h264_decode_frame()的定义可以看出,它调用了decode_nal_units()完成了具体的H.264解码工作。有关H.264解码就不在详细分析了。
给我老师的人工智能教程打call!http://blog.csdn.net/jiangjunshow
![这里写图片描述](https://i-blog.csdnimg.cn/blog_migrate/2e19ffb3303a490d20ea5d00dd397156.gif)