ffmpeg用filter实现视频scale

1、概述
此例子用ffmpeg的filter实现视频scale。
2、代码 
/** 
 * 最简单的基于FFmpeg的AVFilter例子(scale) 
 * 
 * 缪国凯(MK)
 * 821486004@qq.com
 *
 * http://blog.csdn.net/dancing_night
 * 
 * 本程序使用FFmpeg的AVfilter实现了视频的缩放功能。
 * 
 *
 */
 
#include "stdafx.h"
 
#ifdef __cplusplus
extern "C"
{
#endif
#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#include <libavfilter/avfiltergraph.h>
#include <libavfilter/avcodec.h>
#include <libavfilter/buffersink.h>
#include <libavfilter/buffersrc.h>
#include <libavutil/avutil.h>
#include <libswscale/swscale.h>
#ifdef __cplusplus
};
#endif
 
#pragma comment(lib, "avcodec.lib")
#pragma comment(lib, "avformat.lib")
#pragma comment(lib, "avutil.lib")
#pragma comment(lib, "avdevice.lib")
#pragma comment(lib, "avfilter.lib")
 
//#pragma comment(lib, "avfilter.lib")
//#pragma comment(lib, "postproc.lib")
//#pragma comment(lib, "swresample.lib")
#pragma comment(lib, "swscale.lib")
 
 
static AVFormatContext *ifmt_ctx, *ofmt_ctx;
static AVCodecContext *pCodecCtx;
AVFilterContext *buffersink_ctx;
AVFilterContext *buffersrc_ctx;
AVFilterGraph *filter_graph;
static int video_stream_index = -1;
static int64_t last_pts = AV_NOPTS_VALUE;
 
 
static int open_input_file(const char *filename)
{
    int ret;
    AVCodec *dec;
 
    if ((ret = avformat_open_input(&ifmt_ctx, filename, NULL, NULL)) < 0) {
        printf( "Cannot open input file\n");
        return ret;
    }
 
    if ((ret = avformat_find_stream_info(ifmt_ctx, NULL)) < 0) {
        printf( "Cannot find stream information\n");
        return ret;
    }
 
    /* select the video stream */
    ret = av_find_best_stream(ifmt_ctx, AVMEDIA_TYPE_VIDEO, -1, -1, &dec, 0);
    if (ret < 0) {
        printf( "Cannot find a video stream in the input file\n");
        return ret;
    }
    video_stream_index = ret;
    pCodecCtx = ifmt_ctx->streams[video_stream_index]->codec;
 
    /* init the video decoder */
    if ((ret = avcodec_open2(pCodecCtx, dec, NULL)) < 0) 
    {
        printf( "Cannot open video decoder\n");
        return ret;
    }
 
    return 0;
}
 
int openoutputfile(const char* filename, int width, int height)
{
    AVStream *out_stream;
    int ret = 0;
    if ((ret = avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, filename)) < 0)
    {
        printf("can not alloc output context");
        return ret;
    }
    for (int i = 0; i < ifmt_ctx->nb_streams; i++)
    {
        //if the stream is video stream then find the encoder default
        //and set context and open the encoder
        if (ifmt_ctx->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO)
        {
            out_stream = NULL;
            //new a stream
            out_stream = avformat_new_stream(ofmt_ctx, NULL);
            if (!out_stream)
            {
                printf("can not new stream for output");
                return AVERROR_UNKNOWN;
            }
 
            //use default video encoder
            out_stream->codec->codec = avcodec_find_encoder(ofmt_ctx->oformat->video_codec);
                                
            out_stream->codec->codec_type = AVMEDIA_TYPE_VIDEO;
            out_stream->codec->pix_fmt = PIX_FMT_YUV420P;
            out_stream->codec->width = width;  
            out_stream->codec->height = height;
            out_stream->codec->time_base.num = 1;  
            out_stream->codec->time_base.den = 25;  
            out_stream->codec->bit_rate = 400000;  
            out_stream->codec->gop_size=250;
            //H264
            out_stream->codec->qmin = 10;
            out_stream->codec->qmax = 40;
            //Optional Param
            out_stream->codec->max_b_frames=3;
 
            AVDictionary *param = NULL;
            if (out_stream->codec->codec->id == AV_CODEC_ID_H264)
            {
                av_dict_set(¶m, "preset", "slow", 0);
                av_dict_set(¶m, "tune", "zerolatency", 0);
                av_dict_set(¶m, "profile", "main", 0);
            }
 
            //open encoder
            ret = avcodec_open2(out_stream->codec, out_stream->codec->codec, ¶m);
 
            if (ret < 0)
            {
                printf("can not open encoder");
                return ret;
            }
 
            if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
                out_stream->codec->flags |= CODEC_FLAG_GLOBAL_HEADER;
 
            break;
        }
    }
 
    //dump output info
    av_dump_format(ofmt_ctx, 0, filename, 1);
 
    //open the output file handle
    if (!(ofmt_ctx->oformat->flags & AVFMT_NOFILE))
    {
        ret = avio_open(&ofmt_ctx->pb, filename, AVIO_FLAG_WRITE);
        if (ret < 0)
        {
            printf("can not open the output file handle");
            return ret;
        }
    }
 
    //write output file header
    if ((ret = avformat_write_header(ofmt_ctx, NULL)) < 0)
    {
        printf("can not write output file header");
        return ret;
    }
 
    return 0;
}
 
static int init_filters(const char *filters_descr)
{
    char args[512];
    int ret;
    AVFilter *buffersrc  = avfilter_get_by_name("buffer");
    AVFilter *buffersink = avfilter_get_by_name("ffbuffersink");
    AVFilterInOut *outputs = avfilter_inout_alloc();
    AVFilterInOut *inputs  = avfilter_inout_alloc();
    enum PixelFormat pix_fmts[] = { PIX_FMT_YUV420P, PIX_FMT_NONE };
    AVBufferSinkParams *buffersink_params;
 
    filter_graph = avfilter_graph_alloc();
 
    /* buffer video source: the decoded frames from the decoder will be inserted here. */
    _snprintf(args, sizeof(args),
        "video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
        pCodecCtx->width, pCodecCtx->height, pCodecCtx->pix_fmt,
        pCodecCtx->time_base.num, pCodecCtx->time_base.den,
        pCodecCtx->sample_aspect_ratio.num, pCodecCtx->sample_aspect_ratio.den);
 
    ret = avfilter_graph_create_filter(&buffersrc_ctx, buffersrc, "in",
        args, NULL, filter_graph);
    if (ret < 0) {
        printf("Cannot create buffer source\n");
        return ret;
    }
 
    /* buffer video sink: to terminate the filter chain. */
    buffersink_params = av_buffersink_params_alloc();
    buffersink_params->pixel_fmts = pix_fmts;
    ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
        NULL, buffersink_params, filter_graph);
    av_free(buffersink_params);
    if (ret < 0) {
        printf("Cannot create buffer sink\n");
        return ret;
    }
 
    /* Endpoints for the filter graph. */
    outputs->name       = av_strdup("in");
    outputs->filter_ctx = buffersrc_ctx;
    outputs->pad_idx    = 0;
    outputs->next       = NULL;
 
    inputs->name       = av_strdup("out");
    inputs->filter_ctx = buffersink_ctx;
    inputs->pad_idx    = 0;
    inputs->next       = NULL;
 
    if ((ret = avfilter_graph_parse_ptr(filter_graph, filters_descr,
        &inputs, &outputs, NULL)) < 0)
        return ret;
 
    if ((ret = avfilter_graph_config(filter_graph, NULL)) < 0)
        return ret;
    return 0;
}
 
static int encode_write_video_frame(AVFrame *filt_frame, int *got_frame) 
{
    int ret;
    int got_frame_local;
    AVPacket enc_pkt;
    unsigned int stream_index = 0;
    if (!got_frame)
        got_frame = &got_frame_local;
    av_log(NULL, AV_LOG_INFO, "Encoding frame\n");
    /* encode filtered frame */
    enc_pkt.data = NULL;
    enc_pkt.size = 0;
    av_init_packet(&enc_pkt);
    ret = avcodec_encode_video2(ofmt_ctx->streams[stream_index]->codec, &enc_pkt,
        filt_frame, got_frame);
    av_frame_free(&filt_frame);
    if (ret < 0)
        return ret;
    if (!(*got_frame))
        return 0;
    /* prepare packet for muxing */
    enc_pkt.stream_index = stream_index;
    enc_pkt.dts = av_rescale_q_rnd(enc_pkt.dts,
        ofmt_ctx->streams[stream_index]->codec->time_base,
        ofmt_ctx->streams[stream_index]->time_base,
        (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
    enc_pkt.pts = av_rescale_q_rnd(enc_pkt.pts,
        ofmt_ctx->streams[stream_index]->codec->time_base,
        ofmt_ctx->streams[stream_index]->time_base,
        (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
    enc_pkt.duration = av_rescale_q(enc_pkt.duration,
        ofmt_ctx->streams[stream_index]->codec->time_base,
        ofmt_ctx->streams[stream_index]->time_base);
    av_log(NULL, AV_LOG_DEBUG, "Muxing frame\n");
    /* mux encoded frame */
    ret = av_interleaved_write_frame(ofmt_ctx, &enc_pkt);
    return ret;
}
 
static int filter_encode_write_video_frame(AVFrame *frame)
{
    int ret;
    AVFrame *filt_frame;
    av_log(NULL, AV_LOG_INFO, "Pushing decoded frame to filters\n");
    /* push the decoded frame into the filtergraph */
    ret = av_buffersrc_add_frame_flags(buffersrc_ctx,
            frame, 0);
    if (ret < 0) {
        av_log(NULL, AV_LOG_ERROR, "Error while feeding the filtergraph\n");
        return ret;
    }
    /* pull filtered frames from the filtergraph */
    while (1) {
        filt_frame = av_frame_alloc();
        if (!filt_frame) {
            ret = AVERROR(ENOMEM);
            break;
        }
        av_log(NULL, AV_LOG_INFO, "Pulling filtered frame from filters\n");
        ret = av_buffersink_get_frame(buffersink_ctx,
                filt_frame);
        if (ret < 0) {
            /* if no more frames for output - returns AVERROR(EAGAIN)
             * if flushed and no more frames for output - returns AVERROR_EOF
             * rewrite retcode to 0 to show it as normal procedure completion
             */
            if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
                ret = 0;
            av_frame_free(&filt_frame);
            break;
        }
        filt_frame->pict_type = AV_PICTURE_TYPE_NONE;
        ret = encode_write_video_frame(filt_frame, NULL);
        if (ret < 0)
            break;
    }
    return ret;
}
 
int flush_encoder(AVFormatContext *fmt_ctx,unsigned int stream_index)
{
    int ret;
    int got_frame;
    if (!(ofmt_ctx->streams[stream_index]->codec->codec->capabilities &
        CODEC_CAP_DELAY))
        return 0;
    while (1) {
        av_log(NULL, AV_LOG_INFO, "Flushing stream #%u encoder\n", stream_index);
        ret = encode_write_video_frame(NULL, &got_frame);
        if (ret < 0)
            break;
        if (!got_frame)
            return 0;
    }
    return ret;
}
 
int _tmain(int argc, _TCHAR* argv[])
{
    char filter_descr[100]/* = "movie=my_logo.png[wm];[in][wm]overlay=5:5[out]"*/;
    AVPacket pkt_in, pkt_out;
    unsigned int stream_index;
    int ret;
    AVPacket packet;
    AVFrame *frame;
    int got_frame;
    int width, height;
    width = 400;
    height = 300;
    
    sprintf(filter_descr, "[in]scale=%d:%d[out]", width, height);
 
    avcodec_register_all();
    av_register_all();
    avfilter_register_all();
 
    if ((ret = open_input_file("test.mp4")) < 0)
        goto end;
    if ((ret = init_filters(filter_descr)) < 0)
        goto end;
    if ((ret = openoutputfile("test_scale.mp4", width, height)) < 0)
        goto end;
    
    // to be add
    while(1)
    {
        if (av_read_frame(ifmt_ctx, &pkt_in) < 0)
        {
            break;
        }
        pkt_out.data = NULL;
        pkt_out.size = 0;
        av_init_packet(&pkt_out);
        stream_index = pkt_in.stream_index;
        frame = av_frame_alloc();
        int got_frame = -1;
        int ret = -1;
 
        //calculate the pts and dts
        pkt_in.dts = av_rescale_q_rnd(pkt_in.dts,
            ifmt_ctx->streams[stream_index]->time_base,
            ifmt_ctx->streams[stream_index]->codec->time_base,
            (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
        pkt_in.pts = av_rescale_q_rnd(pkt_in.pts,
            ifmt_ctx->streams[stream_index]->time_base,
            ifmt_ctx->streams[stream_index]->codec->time_base,
            (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
 
        
        if (ifmt_ctx->streams[stream_index]->codec->codec_type == AVMEDIA_TYPE_VIDEO)
        {
            ret = avcodec_decode_video2(ifmt_ctx->streams[stream_index]->codec, frame, &got_frame, &pkt_in);
            if (ret < 0)
            {
                av_frame_free(&frame);
                printf("decoding video stream failed\n");
                break;
            }
 
            if (got_frame) 
            {
                frame->pts = av_frame_get_best_effort_timestamp(frame);
                ret = filter_encode_write_video_frame(frame);
                av_frame_free(&frame);
                if (ret < 0)
                    goto end;
            }
 
        }
    }
 
 
    //Flush Encoder
    ret = flush_encoder(ofmt_ctx,0);
    if (ret < 0) {
        printf("Flushing encoder failed\n");
        return -1;
    }
 
    av_write_trailer(ofmt_ctx);
    
end:
    avfilter_graph_free(&filter_graph);
    if (pCodecCtx)
        avcodec_close(pCodecCtx);
    avformat_close_input(&ifmt_ctx);
 
    if (ofmt_ctx && !(ofmt_ctx->oformat->flags & AVFMT_NOFILE))
    {
        for (int i = 0; i < ofmt_ctx->nb_streams; i++)
        {
            if (ofmt_ctx->streams[i]->codec)
            {
                avcodec_close(ofmt_ctx->streams[i]->codec);
            }
        }
 
        avio_close(ofmt_ctx->pb);
    }
    avformat_free_context(ofmt_ctx);
 
    if (ret < 0 && ret != AVERROR_EOF) {
        char buf[1024];
        av_strerror(ret, buf, sizeof(buf));
        printf("Error occurred: %s\n", buf);
        return -1;
    }
 
    return 0;
}
3、解释
简单说下流程:open input->open output->init filter -> read packet -> decode frame -> push into filter -> pull from filter -> encode -> write into file
4、工程下载
http://download.csdn.net/detail/dancing_night/9066603
ref:https://blog.csdn.net/dancing_night/article/details/48131755

上一篇:LOVE


下一篇:“WPF老矣,尚能饭否”—且说说WPF今生未来(上):担心