There are a lot of questions on this topic, and I've read most of them and most of the google search results I could come up with.
When I use FFMPEG to convert a FLV to a iphone3 compatble MP4 file, it just doesn't preserver enough of the quality. Yes, I've worked the hell out of -sameq and -b and -bt settings, text just isn't readable.
Next I tried to split the video out and process it directly, using these instructions:
https://sites.google.com/site/linuxencoding/x264-encoding-guide
The problem is myplayer (via ffmpeg) was not able to determine the duration of the FLV (even though the metadata was set).
(I assume) Because of that unknown duration, when I create the MP4 file, the resulting x264 file plays through super-fast while the audio plays at the normal rate.
user#server:/tmp# mplayer -nosound -benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'
MPlayer 1.0rc4-4.4.5 (C) 2000-2010 MPlayer Team
mplayer: could not connect to socket
mplayer: No such file or directory
Failed to open LIRC support. You will not be able to use your remote control.
Playing input.flv.
libavformat file format detected.
[flv # 0x1202460]Estimating duration from bitrate, this may be inaccurate
[lavf] stream 0: video (vp6f), -vid 0
[lavf] stream 1: audio (nellymoser), -aid 0
VIDEO: [VP6F] 1680x992 0bpp 1000.000 fps 33.4 kbps ( 4.1 kbyte/s)
Clip info:
audiocodecid: 6
audiodatarate: 86
audiosamplerate: 44100
audiosamplesize: 16
audiosize: 6097005
canSeekToEnd: true
datasize: 8609138
duration: 567
framerate: 2
hasAudio: true
hasCuePoints: false
hasKeyframes: true
hasMetadata: true
hasVideo: true
height: 992
lasttimestamp: 567
metadatacreator: flvtool++ (Facebook, Motion project, dweatherford)
stereo: false
totalframes: 1043
videocodecid: 4
videodatarate: 33
videosize: 2316256
width: 1680
Using (default) progressive frame mode.Opening video filter: [expand w=640 h=480]
Expand: 640 x 480, -1 ; -1, osd: 0, aspect: 0.000000, round: 1
Opening video filter: [scale w=0 h=0]
Opening video filter: [dsize=640:480:0]
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffvp6f] vfm: ffmpeg (FFmpeg VP6 Flash)
==========================================================================
Audio: no sound
Starting playback...
Movie-Aspect is undefined - no prescaling applied.
[swscaler # 0x7f0c738b9620]Lanczos scaler, from yuv420p to yuv420p using MMX2
VO: [yuv4mpeg] 640x480 => 641x480 Planar YV12
I have also tried specifying FPS, but no change in results
user#server:/tmp# mplayer -nosound -fps 25-benchmark -sws 9 -vf dsize=640:480:0,scale=0:0,expand=640:480 -vo yuv4mpeg:file=>(x264 --demuxer y4m --fps 25 --crf 0 --preset slow --threads auto --output output.264 - 2>x264.log) 'input.flv'
Can someone tell me how to either:
fix my split A/V processing/timing/duration issues?
improve the
quality of the FFMPEG conversion of FLV to iphone3 compatible
format?
I would suggest trying handbreak(http://handbrake.fr/), I think it can encode flvs as well as almost every other file formate I have ever tried, and I'v never had any trouble with quality.
Related
IMHO, MP4 (ISO/IEC 14496-12) is quite complicated. This question is meant for those familiar with ins and outs of MP4, so I try to be brief. Otherwise, this question could be a long essay.
I am trying to create fMP4 with C# (the language is not relevant here). The created fMP4 file does not play. Using the tool ffprobe to check the file generates the following:
ffprobe "foo.mp4"
[h264 # 0000021625d10e40] no frame!
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'foo.mp4':
Metadata:
major_brand : mp42
minor_version : 0
compatible_brands: avc1mp42isomiso6
creation_time : 2022-09-11T00:24:35.000000Z
Duration: 00:00:10.06, start: 0.000000, bitrate: 2078 kb/s
Stream #0:0(eng): Video: h264 (Main) (avc1 / 0x31637661), yuv420p(tv, bt709), 1280x800 [SAR 1:1 DAR 8:5], 2077 kb/s, 12.22 fps, 30 tbr, 1000k tbn, 2000k tbc (default)
Metadata:
creation_time : 2022-09-11T00:24:35.000000Z
handler_name : Media Handler
vendor_id : [0][0][0][0]
encoder : AVC Coding
Please note the error:
[h264 # 0000021625d10e40] no frame!
When I use:
ffprobe -show_frames "foo.mp4"
It shows hundreds of frames of this small MP4 file that seem flawless. I cannot paste them here there are thousands of lines. However, ffprobe shows two errors at the start:
[h264 # 000001b015110ec0] no frame!
[h264 # 000001b01545e7c0] no frame!
Please note these two "no frame!" errors generated by ffprobe -show_frames are different from the first one by ffprobe without the -show_frames option.
I wonder if anyone could offer a tip about what these "no frame!" errors mean and where I should chase the culprit.
The following is what Mp4 Explorer shows:
Update 2022-09-17
Here is an fMP4 file: example.mp4. Every video player seems to be able to play it except for VLC Player. VLC Player is an important player. I would love to make it work for it. Could anyone offer a clue about why VLC Player does not like it?
I am trying to capture the canvas on browser (canvas.captureStream), add audio ( captured using getUserMedia and added to stream using canvasStream.addTracks) and send it to server. Server sends the stream ( after encoding with H264 ) to facebook live using ffmpeg. However the stream is not stable and gets disconnected within minutes.
If the video along with audio tracks (using getMediaUser) is directly(without canvas.captureStream) sent it works fine. I am suspecting it has to do with the audio track not being proper and hence facebook rejecting it ( no particular error from ffmpeg, it just exits with IO error). Need help in figuring out the right way to send canvas stream along with audio.
Current ffmpeg command is as below
ffmpeg -i - -c:v libx264 -crf 23 -preset ultrafast \
-tune zerolatency -max_muxing_queue_size 1000 \
-vsync cfr -async 1 -bufsize 2M -r 30 -g 60 -keyint_min 30 \
-x264opts keyint=30 -pix_fmt yuv420p -level 3 \
-c:a aac -b:a 96k -ar 96000 \
-f tee -map 0:v -map 0:a [f=flv] <rtmp url 1> | [f=flv] <rtmp url 2>
Note: In the above command <rtmp url 1> and <rtmp url 2> are replaced with actual urls ( including the session key required )
My settings to capture canvas and audio on browser are as below
canvasStream = canvas.captureStream(30);
audioTrack = stream.getTracks().filter( (track) => {
return track.kind === 'audio';
})[0];
canvasStream.addTrack(audioTrack);
Note: Audio constraints given in getUserMedia are as below
audio: {
sampleRate: 44100,
echoCancellation: true
}
Error given by ffmpeg while exiting is as below
FFMPEG:[flv # 0x5626ecf0f7c0] Failed to update header with correct duration.
[flv # 0x5626ecf0f7c0] Failed to update header with correct filesize.
FFMPEG:[tee # 0x5626ec8550c0] Slave muxer #1 failed, aborting.
av_interleaved_write_frame(): Broken pipe
FFMPEG:[flv # 0x5626ecbe8840] Failed to update header with correct duration.
FFMPEG:[flv # 0x5626ecbe8840] Failed to update header with correct filesize.
FFMPEG:frame= 2378 fps= 22 q=22.0 Lsize=N/A time=00:01:21.17 bitrate=N/A dup=2202 drop=5 speed=0.745x
video:18916kB audio:959kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
FFMPEG:[libx264 # 0x5626ec855b80] frame I:80 Avg QP:13.97 size:115968
[libx264 # 0x5626ec855b80] frame P:2298 Avg QP:16.77 size: 4392
[libx264 # 0x5626ec855b80] mb I I16..4: 100.0% 0.0% 0.0%
FFMPEG:[libx264 # 0x5626ec855b80] mb P I16..4: 3.8% 0.0% 0.0% P16..4: 9.5% 0.0% 0.0% 0.0% 0.0% skip:86.7%
[libx264 # 0x5626ec855b80] coded y,uvDC,uvAC intra: 26.3% 42.2% 13.6% inter: 2.9% 6.1% 0.1%
[libx264 # 0x5626ec855b80] i16 v,h,dc,p: 34% 26% 23% 17%
[libx264 # 0x5626ec855b80] i8c dc,h,v,p: 46% 27% 20% 7%
[libx264 # 0x5626ec855b80] kb/s:1954.96
[aac # 0x5626ec84d200] Qavg: 122.161
Conversion failed!
Appreciate any help with this.
I am suspecting it has to do with the audio track not being proper and hence facebook rejecting it
Yes, this is correct. Facebook requires an audio track and will drop your stream if it isn't present.
For starters, drop -ar 96000 from your FFmpeg command. This would be setting a sample rate of 96 kHz, which Facebook isn't going to want. You really don't want to resample at FFmpeg anyway... just let it use whatever sample rate it gets from the browser.
Next, set your audio sample rate to 48 kHz rather than 44.1 kHz on your getUserMedia constraints. Facebook doesn't handle 44.1 kHz audio on the RTMP ingest well. (Generally these days, you want to use 48 kHz for everything unless you're targeting CD audio, which would be 44.1 kHz.)
Next, your FFmpeg audio command is outputting to two RTMP URLs. If one fails, they will both fail. If one gets behind, they will both be stalled. This is probably not what you want. :-) Drop one of them.
Now, make sure you have a decent and consistent frame rate. Facebook is very picky about this and will drop your stream.
To test/debug, replace your RTMP URL output with a simple FLV file on disk, to ensure things are working from the browser up through FFmpeg. Once you have that all smooth and working, you can reconnect to Facebook and go from there.
I've been able to record .h264 files without a problem, both using raspivid and through Python, but whatever I try, I can't convert them to a .mp4, .mkv or .avi file.
I've tried converting using MP4box (suggested in several places) and mkvmerge, but to no avail.
When I use MP4box (MP4box -add <source.h264> <dest.mp4>), I get :
AVC-H264 import - frame size 1920 x 1080 at 25.000 FPS
WARNING: NAL Unit type 0 not handled - adding5/100)
AVC Import results: 44 samples - Slices: 3 I 41 P 0 B - 0 SEI - 3 IDR
Saving video.mp4: 0.500 secs Interleaving
Then I try to play this file using VLC, it doesn't give an error, but shows garbled data.
To be clear, the camera works fine (shows perfect video using Python code and raspistill outputs a perfect jpg file). I tried a different camera as well.
An attempt with mkvmerge :
pi#raspberrypi:~/cam $ raspivid -o test.h264 -fps 30 -t 15000 -w 1920 -h 1080
pi#raspberrypi:~/cam $ mkvmerge --default-duration 0:30p -o video.mkv test.h264
mkvmerge v31.0.0 ('Dolores In A Shoestand') 32-bit
'test.h264': Using the demultiplexer for the format 'AVC/h.264'.
'test.h264' track 0: Using the output module for the format 'AVC/h.264 (unframed)'.
The file 'video.mkv' has been opened for writing.
Error: 'test.h264' track 0: mkvmerge encountered broken or unparsable data in this AVC/h.264 video track. Either your file is damaged (which mkvmerge cannot cope with yet) or this is a bug in mkvmerge itself. The error message was:
Success
pi#raspberrypi:~/cam $
This is on a Raspberry Pi 4.
A sample file can be found here : https://filebin.net/c40usz0crhgggadf
Created with : raspivid -t 30000 -w 640 -h 480 -fps 25 -b 1200000 -p 0,0,640,480 -o pivideo.h264
I'm going to respond to this myself. After doing an 'apt-get upgrade', the problem is gone.
If anyone wants to figure this out to find the root cause, this is what apt-get did :
The following packages will be upgraded:
bind9-host bluez-firmware firmware-atheros firmware-brcm80211 firmware-libertas firmware-misc-nonfree firmware-realtek libbind9-161 libdns-export1104 libdns1104 libgs9 libgs9-common libisc-export1100 libisc1100 libisccc161
libisccfg163 liblwres161 libraspberrypi-bin libraspberrypi-dev libraspberrypi-doc libraspberrypi0 pi-bluetooth raspberrypi-bootloader raspberrypi-kernel raspberrypi-kernel-headers raspi-config rpd-plym-splash
The following packages will be DOWNGRADED:
python3-pgzero
27 upgraded, 0 newly installed, 1 downgraded, 0 to remove and 0 not upgraded.
I have been struggling to get VideoReader working on Ubuntu 17.04. If I run a simple program:
v = VideoReader('flame.avi')
I get the following:
Error using VideoReader/init (line 619)
Could not read file due to an unexpected error. Reason: Unable to initialize the video obtain properties
Error in VideoReader (line 172)
obj.init(fileName);
Now, I have done a little bit of digging, and most people say that this error comes from not having gstreamer0.10 installed.
See: https://www.mathworks.com/matlabcentral/answers/165584-cannot-read-video-matlab-2014b-linux
This one says that gstreamer1.0 is not sufficient to fix this problem: https://www.mathworks.com/matlabcentral/answers/255157-can-t-load-video-using-videoreader-in-matlab-r2015-on-ubuntu-15-04-could-not-read-file-due-to-an-un
The problem comes from trying to install gstreamer0.10. It doesn't appear to be working in Ubuntu 17.
This one talks about how gstreamer isn't working on Ubuntu 17: https://www.mathworks.com/matlabcentral/answers/294258-hello-i-have-a-problem-with-videoreader-on-matlab-r2016a-with-ubtuntu-lts16-04
I was wondering if anyone had any ideas on how to overcome this issue? Is there a different package that I could install that would fulfill the same requirements? I have tried installing matlab-support, and that hasn't helped either. Is there a different issue other than gstreamer0.10?
I would prefer not to re-install Ubuntu 16.04, but understand if that is the only way. But surely, someone has gotten MATLAB working on Ubuntu 17?
A colleague of mine found a link to a post that solved my problem https://www.mathworks.com/matlabcentral/answers/329796-issue-with-libstdc-so-6
Essentially, you need to redirect MATLAB and tell it to not use its default libstdc++6 file and use your Linux system's instead. I went with the solution at the bottom, and wrote an alias for my .bashrc file.
alias matlab='LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 /usr/local/bin/matlab -desktop'
After testing it out, I have confirmed it works.
I had the exact same problem and I was able to solve it by running
sudo apt install gstreamer1.0-plugins-*
on Ubuntu 20.04 with MATLAB R2018a
Moreover, forcing MATLAB to use the system libstdc++ in ways 1. and 2. did not solve the problem:
Write in ~/.bashrc
alias matlab='LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 /usr/local/bin/matlab -desktop'
closing the terminal, open again, run matlab and still did not work.
going to /usr/local/MATLAB/R2018a/sys/os/glnxa64 and adding .bak to all libstdc++ files (as a means of making them "invisible" to matlab and force it to look for them in the system) did not work either.
EDIT
This is working for at least one video, but still other videos give me the same error. So it is not a full solution
EDIT
One such video I could not open with matlab's video reader becomes accessible after converting it with this script (using opencv).
import cv2
import tqdm
cap = cv2.VideoCapture("output.avi")
vw = None
nframes = 100
# for fast test, change the for loop to
# while True:
# to convert the whole video
for _ in tqdm.tqdm(range(nframes)):
ret, frame = cap.read()
if not ret:
break
if vw is None:
vw = cv2.VideoWriter("output_safe.avi", cv2.VideoWriter_fourcc('M','J','P','G'), 10, (frame.shape[1], frame.shape[0]))
vw.write(frame)
cap.release()
if vw is None:
vw.release()
These were the properties of the original video as reported with ffmpeg -i output.avi
Input #0, avi, from 'output.avi':
Metadata:
encoder : Lavf58.64.100
Duration: 00:01:00.00, start: 0.000000, bitrate: 1822 kb/s
Stream #0:0: Video: mpeg4 (Simple Profile) (FMP4 / 0x34504D46), yuv420p, 2592x1944 [SAR 1:1 DAR 4:3], 1817 kb/s, 25 fps, 25 tbr, 25 tbn, 25 tbc
and these are the new properties
Input #0, avi, from 'output_safe.avi':
Metadata:
encoder : Lavf58.61.100
Duration: 00:00:10.00, start: 0.000000, bitrate: 7594 kb/s
Stream #0:0: Video: mjpeg (Baseline) (MJPG / 0x47504A4D), yuvj420p(pc, bt470bg/unknown/unknown), 2592x1944, 7650 kb/s, 10 fps, 10 tbr, 10 tbn, 10 tbc
For an app I'm developing, I need avconv to stream from a link for a specified length of time. I've tried using -timelimit to no avail. What am I doing wrong? Is there a better way to do this? Why doesn't -timelimit work?
See, for example, the call below, which runs until terminated with ctrl-c.
stack#ThinkPad:~/app_dev$ avconv -timelimit 30 -i http://br-mp3-bayern2sued-m.akacast.akamaistream.net/7/731/256282/v1/gnl.akacast.akamaistream.net/br_mp3_bayern2sued_m stack_test.mp3
Which generates the following output.
avconv version 9.18-6:9.18-0ubuntu0.14.04.1, Copyright (c) 2000-2014 the Libav developers
built on Mar 16 2015 13:19:10 with gcc 4.8 (Ubuntu 4.8.2-19ubuntu1)
[mp3 # 0x765100] Header missing
[mp3 # 0x75f020] max_analyze_duration reached
[mp3 # 0x75f020] Estimating duration from bitrate, this may be inaccurate
Guessed Channel Layout for Input Stream #0.0 : stereo
Input #0, mp3, from 'http://br-mp3-bayern2sued-m.akacast.akamaistream.net/7/731/256282/v1/gnl.akacast.akamaistream.net/br_mp3_bayern2sued_m':
Duration: N/A, start: 0.000000, bitrate: 128 kb/s
Stream #0.0: Audio: mp3, 44100 Hz, stereo, s16p, 128 kb/s
Output #0, mp3, to 'stack_test.mp3':
Metadata:
TSSE : Lavf54.20.4
Stream #0.0: Audio: libmp3lame, 44100 Hz, stereo, s16p
Stream mapping:
Stream #0:0 -> #0:0 (mp3 -> libmp3lame)
Press ctrl-c to stop encoding
[mp3 # 0x765100] Header missing
Error while decoding stream #0:0
[mp3 # 0x7639e0] Application provided invalid, non monotonically increasing dts to muxer in stream 0: -2255 >= -2255
-t and -timelimit are different. More importantly, you need to use -t as an output option, not as an input option. Updating your example, the following should work (tested on ffmpeg).
ffmpeg -i http://br-mp3-bayern2sued-m.akacast.akamaistream.net/7/731/256282/v1/gnl.akacast.akamaistream.net/br_mp3_bayern2sued_m -t 30 stack_test.mp3