I am attempting to encode a mp4 video file into a flv video. Here is what I am doing:
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <math.h>
#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#define STREAM_FRAME_RATE 30 /* 30 images/s */
#define STREAM_PIX_FMT AV_PIX_FMT_YUV420P /* default pix_fmt */
/**************************************************************/
/* Add an output stream. */
static AVStream *add_stream(AVFormatContext *oc, AVCodec **codec,
enum AVCodecID codec_id) {
AVCodecContext *c;
AVStream *st;
/* find the encoder */
*codec = avcodec_find_encoder(codec_id);
if (!(*codec)) {
fprintf(stderr, "Could not find encoder for '%s'\n",
avcodec_get_name(codec_id));
exit(1);
}
st = avformat_new_stream(oc, *codec);
if (!st) {
fprintf(stderr, "Could not allocate stream\n");
exit(1);
}
st->id = oc->nb_streams - 1;
c = st->codec;
switch ((*codec)->type) {
case AVMEDIA_TYPE_VIDEO:
//avcodec_get_context_defaults3(c, *codec);
c->codec_id = codec_id;
c->bit_rate = 150 * 1000; //2314000;;
//c->rc_max_rate = 150*1000;
//c->rc_buffer_size = 150*1000;
/* Resolution must be a multiple of two. */
c->width = 1280;
c->height = 720;
/* timebase: This is the fundamental unit of time (in seconds) in terms
* of which frame timestamps are represented. For fixed-fps content,
* timebase should be 1/framerate and timestamp increments should be
* identical to 1. */
c->time_base.den = STREAM_FRAME_RATE;
c->time_base.num = 1;
c->gop_size = 12; /* emit one intra frame every twelve frames at most */
c->pix_fmt = STREAM_PIX_FMT;
if (c->codec_id == AV_CODEC_ID_MPEG2VIDEO) {
/* just for testing, we also add B frames */
c->max_b_frames = 2;
}
if (c->codec_id == AV_CODEC_ID_MPEG1VIDEO) {
/* Needed to avoid using macroblocks in which some coeffs overflow.
* This does not happen with normal video, it just happens here as
* the motion of the chroma plane does not match the luma plane. */
c->mb_decision = 2;
}
break;
default:
break;
}
/* Some formats want stream headers to be separate. */
if (oc->oformat->flags & AVFMT_GLOBALHEADER)
c->flags |= CODEC_FLAG_GLOBAL_HEADER;
return st;
}
/**************************************************************/
/* video output */
static AVFrame *frame;
static AVPicture src_picture, dst_picture;
static int frame_count;
static void open_video(AVFormatContext *oc, AVCodec *codec, AVStream *st) {
int ret;
AVCodecContext *c = st->codec;
/* open the codec */
ret = avcodec_open2(c, codec, NULL);
if (ret < 0) {
fprintf(stderr, "Could not open video codec: %s\n", av_err2str(ret));
exit(1);
}
/* allocate and init a re-usable frame */
frame = avcodec_alloc_frame();
if (!frame) {
fprintf(stderr, "Could not allocate video frame\n");
exit(1);
}
/* Allocate the encoded raw picture. */
ret = avpicture_alloc(&dst_picture, c->pix_fmt, c->width, c->height);
if (ret < 0) {
fprintf(stderr, "Could not allocate picture: %s\n", av_err2str(ret));
exit(1);
}
/* If the output format is not YUV420P, then a temporary YUV420P
* picture is needed too. It is then converted to the required
* output format. */
if (c->pix_fmt != AV_PIX_FMT_YUV420P) {
ret = avpicture_alloc(&src_picture, AV_PIX_FMT_YUV420P, c->width,
c->height);
if (ret < 0) {
fprintf(stderr, "Could not allocate temporary picture: %s\n",
av_err2str(ret));
exit(1);
}
}
/* copy data and linesize picture pointers to frame */
*((AVPicture *) frame) = dst_picture;
}
static void close_video(AVFormatContext *oc, AVStream *st) {
avcodec_close(st->codec);
av_free(src_picture.data[0]);
av_free(dst_picture.data[0]);
av_free(frame);
}
int main(int argc, char *argv[]) {
AVFormatContext *pFormatCtx = NULL;
int i, videoStream;
AVCodecContext *pCodecCtx = NULL;
AVCodec *pCodec;
AVFrame *pFrame;
AVPacket packet;
int frameFinished;
const char *filename;
AVOutputFormat *fmt;
AVFormatContext *oc;
AVStream *video_st;
AVCodec *video_codec;
int ret;
// Register all formats, codecs and network
av_register_all();
avcodec_register_all();
avformat_network_init();
// Open video file
if (avformat_open_input(&pFormatCtx, "input_file.mp4", NULL, NULL) != 0)
return -1; // Couldn't open file
// Retrieve stream information
if (avformat_find_stream_info(pFormatCtx, NULL) < 0)
return -1; // Couldn't find stream information
// Dump information about file onto standard error
av_dump_format(pFormatCtx, 0, "input_file.mp4", 0);
// Find the first video stream
videoStream = -1;
for (i = 0; i < pFormatCtx->nb_streams; i++)
if (pFormatCtx->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO) {
videoStream = i;
break;
}
if (videoStream == -1)
return -1; // Didn't find a video stream
// Get a pointer to the codec context for the video stream
pCodecCtx = pFormatCtx->streams[videoStream]->codec;
// Find the decoder for the video stream
pCodec = avcodec_find_decoder(pCodecCtx->codec_id);
if (pCodec == NULL) {
fprintf(stderr, "Unsupported codec!\n");
return -1; // Codec not found
}
// Open codec
if (avcodec_open2(pCodecCtx, pCodec, NULL) < 0)
return -1; // Could not open codec
// Allocate video frame
pFrame = avcodec_alloc_frame();
// setup mux
filename = "output_file.flv";
fmt = av_guess_format("flv", filename, NULL);
if (fmt == NULL) {
printf("Could not guess format.\n");
return -1;
}
// allocate the output media context
oc = avformat_alloc_context();
if (oc == NULL) {
printf("could not allocate context.\n");
return -1;
}
oc->oformat = fmt;
// Add the video streams using the default format codecs
// and initialize the codecs.
video_st = NULL;
if (fmt->video_codec != AV_CODEC_ID_NONE) {
video_st = add_stream(oc, &video_codec, fmt->video_codec);
}
// Now that all the parameters are set, we can open the
// video codec and allocate the necessary encode buffers.
if (video_st)
open_video(oc, video_codec, video_st);
/* open the output file, if needed */
if (!(fmt->flags & AVFMT_NOFILE)) {
ret = avio_open(&oc->pb, filename, AVIO_FLAG_WRITE);
if (ret < 0) {
fprintf(stderr, "Could not open '%s': %s\n", filename,
av_err2str(ret));
return 1;
}
}
// dump output format
av_dump_format(oc, 0, filename, 1);
// Write the stream header, if any.
ret = avformat_write_header(oc, NULL);
if (ret < 0) {
fprintf(stderr, "Error occurred when opening output file: %s\n",
av_err2str(ret));
return 1;
}
// Read frames, decode, and re-encode
while (av_read_frame(pFormatCtx, &packet) >= 0) {
// Is this a packet from the video stream?
if (packet.stream_index == videoStream) {
// Decode video frame
avcodec_decode_video2(pCodecCtx, pFrame, &frameFinished, &packet);
// Did we get a video frame?
if (frameFinished) {
AVFrame* newFrame = avcodec_alloc_frame(); // Initialize a new frame
int size = avpicture_get_size(video_st->codec->pix_fmt,
video_st->codec->width, video_st->codec->height);
uint8_t* picture_buf = av_malloc(size);
avpicture_fill((AVPicture *) newFrame, picture_buf,
video_st->codec->pix_fmt, video_st->codec->width,
video_st->codec->height);
// Copy only the frame content without any other disturbing
av_picture_copy((AVPicture*) newFrame, (AVPicture*) pFrame,
video_st->codec->pix_fmt, video_st->codec->width,
video_st->codec->height);
// encode the image
AVPacket pkt;
int got_output;
av_init_packet(&pkt);
pkt.data = NULL; // packet data will be allocated by the encoder
pkt.size = 0;
/* Setting newFrame->pts this way produces the error 'non-strictly-monotonic PTS' */
//if(newFrame->pts != AV_NOPTS_VALUE)
//newFrame->pts = av_rescale_q(newFrame->pts, video_st->time_base, video_st->codec->time_base);
// Setting newFrame->pts this does not produce 'non-strictly-monotonic PTS'
newFrame->pts = frame_count;
ret = avcodec_encode_video2(video_st->codec, &pkt, newFrame, &got_output);
if (ret < 0) {
fprintf(stderr, "Error encoding video frame: %s\n", av_err2str(ret));
exit(1);
}
if (got_output) {
if (video_st->codec->coded_frame->key_frame)
pkt.flags |= AV_PKT_FLAG_KEY;
pkt.stream_index = video_st->index;
if (pkt.pts != AV_NOPTS_VALUE)
pkt.pts = av_rescale_q(pkt.pts,
video_st->codec->time_base,
video_st->time_base);
if (pkt.dts != AV_NOPTS_VALUE)
pkt.dts = av_rescale_q(pkt.dts,
video_st->codec->time_base,
video_st->time_base);
// Write the compressed frame to the media file.
ret = av_interleaved_write_frame(oc, &pkt);
} else {
ret = 0;
}
if (ret != 0) {
fprintf(stderr, "Error while writing video frame: %s\n",
av_err2str(ret));
exit(1);
}
fprintf(stderr, "encoded frame #%d\n", frame_count);
frame_count++;
av_free(picture_buf);
av_free_packet(&pkt);
}
}
// Free the packet that was allocated by av_read_frame
av_free_packet(&packet);
}
/* Write the trailer, if any. The trailer must be written before you
* close the CodecContexts open when you wrote the header; otherwise
* av_write_trailer() may try to use memory that was freed on
* av_codec_close(). */
av_write_trailer(oc);
/* Close video encoder (codec). */
if (video_st)
close_video(oc, video_st);
// Free the streams.
for (i = 0; i < oc->nb_streams; i++) {
av_freep(&oc->streams[i]->codec);
av_freep(&oc->streams[i]);
}
if (!(fmt->flags & AVFMT_NOFILE))
/* Close the output file. */
avio_close(oc->pb);
/* free the stream */
av_free(oc);
// Free the YUV frame holding decoded frame
av_free(pFrame);
// Close the decoder (codec)
avcodec_close(pCodecCtx);
// Close the input video file
avformat_close_input(&pFormatCtx);
return 0;
}
If I change the output format to an mp4, mov, or f4v by changing:
// setup mux (FLV OUTPUT)
filename = "output_file.flv";
fmt = av_guess_format("flv", filename, NULL);
if (fmt == NULL) {
printf("Could not guess format.\n");
return -1;
}
to
// setup mux (MP4 OUTPUT)
filename = "output_file.mp4";
fmt = av_guess_format("mp4", filename, NULL);
if (fmt == NULL) {
printf("Could not guess format.\n");
return -1;
}
Then the bitrate is honored. Here is the output of running with a mp4 output
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input_file.mp4':
Metadata:
major_brand : mp42
minor_version : 0
compatible_brands: isomavc1mp42
creation_time : 2010-02-09 19:11:10
Duration: 00:04:31.80, start: 0.000000, bitrate: 2314 kb/s
Stream #0:0(und): Audio: aac (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 125 kb/s
Metadata:
creation_time : 2010-02-09 19:11:10
handler_name : (C) 2007 Google Inc. v08.13.2007.
Stream #0:1(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1280x720 [SAR 1:1 DAR 16:9], 2186 kb/s, 30 fps, 30 tbr, 30k tbn, 60 tbc
Metadata:
creation_time : 2010-02-09 19:11:10
handler_name : (C) 2007 Google Inc. v08.13.2007.
[libx264 @ 02268b80] using cpu capabilities: MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
[libx264 @ 02268b80] profile High, level 3.1
[libx264 @ 02268b80] 264 - core 129 r2230 1cffe9f - H.264/MPEG-4 AVC codec - Copyleft 2003-2012 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deb
lock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chro
ma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1
b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=12 keyint_min=1 scenecut=40 intra_refresh=0 rc_lookahead=12 rc=abr mbtree=1 bitrate=150 ratetol=1.0 qco
mp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to 'output_file.mp4':
Stream #0:0: Video: h264, yuv420p, 1280x720, q=-1--1, 150 kb/s, 90k tbn, 30 tbc
encoded frame #0
encoded frame #1
encoded frame #2
encoded frame #3
.
.
.
encoded frame #8142
encoded frame #8143
encoded frame #8144
encoded frame #8145
[libx264 @ 02174240] frame I:771 Avg QP:47.71 size: 2739
[libx264 @ 02174240] frame P:5083 Avg QP:50.05 size: 833
[libx264 @ 02174240] frame B:2264 Avg QP:48.16 size: 91
[libx264 @ 02174240] consecutive B-frames: 59.6% 6.4% 9.0% 24.9%
[libx264 @ 02174240] mb I I16..4: 34.0% 65.8% 0.2%
[libx264 @ 02174240] mb P I16..4: 6.8% 3.0% 0.0% P16..4: 8.5% 0.4% 1.7% 0.0% 0.0% skip:79.6%
[libx264 @ 02174240] mb B I16..4: 0.0% 0.0% 0.0% B16..8: 1.8% 0.0% 0.0% direct: 0.0% skip:98.2% L0:23.6% L1:76.4% BI: 0.0%
[libx264 @ 02174240] final ratefactor: 57.62
[libx264 @ 02174240] 8x8 transform intra:52.1% inter:98.5%
[libx264 @ 02174240] coded y,uvDC,uvAC intra: 3.1% 14.0% 0.2% inter: 0.1% 0.4% 0.0%
[libx264 @ 02174240] i16 v,h,dc,p: 69% 23% 4% 4%
[libx264 @ 02174240] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 23% 6% 68% 1% 1% 1% 1% 0% 0%
[libx264 @ 02174240] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 41% 18% 18% 4% 4% 5% 3% 3% 3%
[libx264 @ 02174240] i8c dc,h,v,p: 98% 1% 1% 0%
[libx264 @ 02174240] Weighted P-Frames: Y:2.1% UV:1.2%
[libx264 @ 02174240] ref P L0: 72.4% 0.8% 19.1% 7.6% 0.1%
[libx264 @ 02174240] ref B L0: 89.8% 9.1% 1.2%
[libx264 @ 02174240] ref B L1: 96.2% 3.8%
[libx264 @ 02174240] kb/s:193.62
The final file size of output_file.mp4 is about 6.5MB and using ffplay, I notice that bitrate was honored. Of course the quality itself shows.
Here is the output of running with a flv output:
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input_file.mp4':
Metadata:
major_brand : mp42
minor_version : 0
compatible_brands: isomavc1mp42
creation_time : 2010-02-09 19:11:10
Duration: 00:04:31.80, start: 0.000000, bitrate: 2314 kb/s
Stream #0:0(und): Audio: aac (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 125 kb/s
Metadata:
creation_time : 2010-02-09 19:11:10
handler_name : (C) 2007 Google Inc. v08.13.2007.
Stream #0:1(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1280x720 [SAR 1:1 DAR 16:9], 2186 kb/s, 30 fps, 30 tbr, 30k tbn, 60 tbc
Metadata:
creation_time : 2010-02-09 19:11:10
handler_name : (C) 2007 Google Inc. v08.13.2007.
Output #0, flv, to 'output_file.flv':
Stream #0:0: Video: flv1, yuv420p, 1280x720, q=2-31, 150 kb/s, 90k tbn, 30 tbc
encoded frame #0
encoded frame #1
encoded frame #2
encoded frame #3
.
.
.
encoded frame #8142
encoded frame #8143
encoded frame #8144
encoded frame #8145
I noticed that the bitrate is honored as long as the container is a mp4, mov, f4v, or something similar. If I try flv, wmv, or other different containers, then the bitrate is not honored. Also, the bitrate that is automatically set for wmv is different from that of flv. I am not sure what I am doing wrong. I suspect it has to do with the pts, however, I cannot find a solution :(.
H.264has very good compression algorithm and can acheive very low bitrates, but the same cannot be said about other older codecs likeflv1. Its not the container that is causing this difference. Its the videos codec,mp4in ffmpeg usesh.264by default and flv usedflv1(h.263). The containerflvsupporth.264too. and if it is only the container that you are interested in, set vcodec toh.264orlibx264and you will get similar results asmp4.