MP4Box Tool

Content

Getting MP4Box Binaries

Commands

    Getting General Info on mp4-file

    Encapsulation h264 elementary stream into mp4 format 

    Concatenation of mp4 files into new mp4-file

    In-Place Removal Hint Tracks

    MPEG-DASH Segmentation

    How Add Copyright In-Place

Add Language Tag

Extract Elementary Stream

 

 

Getting MP4Box Binaries

MP4Box tool is an alternative to ffmpeg (for mp4 issues), and it’s a part of GPAC package.

For Windows:

Windows version of MP4Box.exe can be downloaded via over-advertised site https://softradar.com/mp4box/download/ , actually you need to extract mp4box.exe and two dll-files: msvcr100.dll and js32.dll.

Alternatively you can download the latest version of GPAC from https://gpac.wp.imt.fr/downloads/gpac-nightly-builds/ , this package contains MP4Box inside, you need check MP4Box to get MP4Box.exe only

 

mp4box.exe  will be located at C:\Program Files\GPAC  and the path environment is updated respectively.

 

For Linux:

sudo apt-get install gpac

 

Commands

Getting General Info on mp4-file

the switch ‘-info’ provides general info and info for each track:

mp4box -info Test_720x576_1200kbps.mp4

* Movie Info *
Timescale 1000 – Duration 00:01:53.082
Fragmented File no – 2 track(s)
File Brand isom – version 512
Created: UNKNOWN DATE
File has no MPEG4 IOD/OD

iTunes Info:
Encoder Software: Lavf58.35.100

Track # 1 Info – TrackID 1 – TimeScale 30000 – Duration 00:01:53.046
Media Info: Language “Undetermined” – Type “vide:avc1” – 3388 samples
Visual Track layout: x=0 y=0 width=720 height=576
MPEG-4 Config: Visual Stream – ObjectTypeIndication 0x21
AVC/H264 Video – Visual Size 720 x 576
AVC Info: 1 SPS – 1 PPS – Profile Baseline @ Level 3.1
NAL Unit length bits: 32
Pixel Aspect Ratio 1:1 – Indicated track size 720 x 576
Self-synchronized

Track # 2 Info – TrackID 2 – TimeScale 44100 – Duration 00:01:53.081
Media Info: Language “Undetermined” – Type “soun:mp4a” – 4870 samples
MPEG-4 Config: Audio Stream – ObjectTypeIndication 0x40
MPEG-4 Audio AAC LC – 2 Channel(s) – SampleRate 44100
Synchronized on stream 1
Alternate Group ID 1

 

One can see that the track 1 is  h264 video (vide:avc1) and the track 2 is audio data (AAC format).

Note: To get metadata related to the track #N use the following command:  MP4Box -info N mp4-file

For example in order to get info about track #1:

       MP4Box -info 1 output.mp4

Output:
Track # 1 Info - TrackID 1 - TimeScale 25000 - Media Duration 00:04:19.600
Track is present in Root OD
Media Info: Language "Undetermined" - Type "vide:avc1" - 6490 samples
Handler name: ETI ISO Video Media Handler
Visual Track layout: x=0 y=0 width=1280 height=720
MPEG-4 Config
    Visual Stream - ObjectTypeIndication 0x21
    AVC/H264 Video - Visual Size 1280 x 720
    AVC Info: 1 SPS - 1 PPS - Profile High @ Level 3.1
    NAL Unit length bits: 32
    Pixel Aspect Ratio 1:1 - Indicated track size 1280 x 720
    Chroma format 1 - Luma bit depth 8 - chroma bit depth 8
    Decoding Buffer size 120906 - Average bitrate 1804 kbps - Max Bitrate 4075 kbps
    No stream dependencies for decoding
    StreamPriority 0

Computed info from media:
    Total size 58561532 bytes - Total samples duration 259560 ms
    Average rate 1804 kbps - Max Rate 3608 kbps
    Chunk durations: min 40 ms - max 40 ms - average 40 ms
    Chunk sizes (bytes): min 183 - max 120906 - average 9023

 

Encapsulation h264 elementary stream into mp4 format 

Encapsulate h264 elementary stream with the frame rate 50fps (if the frame rate is not explicitly mentioned then the default 25 fps is taken):

mp4box -nodrop -add crowdrun1080p.h264:fps=50  -new crowd.mp4

-nodrop  force constant FPS when importing video

-new  – create a new file

Sometime it’s required to change some parameters in mp4 metadata, e.g. the language or track handler name. For example, to change the language into ‘rus’ (Russian) within mdhd-box and the handler name in hdlr-box use:

mp4box -add test.h264:lang=ru:name=”AVC Shevach Video”  -new output.mp4

 

 

 

Concatenation of mp4 files into new mp4-file

the parameter ‘-cat’ should be used two concatenate two mp4-file in a single mp4 file:

mp4box -cat test_001.mp4 -cat test_002.mp4 -new test_12.mp4

 

In-Place Removal Hint Tracks

If mp4-file contains hint tracks you can remove these tracks in-place by the command ‘-unhint’

mp4box -unhint  test.mp4

 

MPEG-DASH Segmentation

To segment the input mp4 file into fragments of m4s files, where each fragment has duration of 4s (depending on key frames)

mp4box -dash 4000 -frag 4000 -rap -segment-name  slava_    input.mp4

 

-dash 4000   segment into 4000ms chunks

-frag 4000  segments are not fragmented further since the duration coincide

-rap forces segments to start random access points, i.e. keyframes. In such a case segment’s duration may differ from 4000ms depending on distribution of key frames.

-segment-name specifies the pattern of segments names. So in this case, the segments will be named like this: slava_1.m4s, slava_2.m4s, …

 

How Add Copyright In-Place

There is a special box ‘cprt’ to put the copyright. This box is signaled within ‘udta’ box.

To add a copyright name in-place use ‘-cprt string’:

mp4box -cprt “Shevach Riabtsev” input.mp4

 

 

Add Language Tag

Language tags are specified by ISO 639-2. Let’s suppose you need set the language tag English (‘en’ according to ISO 639-2) in audio trak #2 of mp4-file (or mov-file) in-place:

 mp4box -lang 2=en  test.mp4

To find audio track (handler type ‘soun’) use ffprobe tool, e.g.:

ffprobe -hide_banner -i test_real_clock.mp4 -show_streams

 

[STREAM]
index=0
codec_name=h264
codec_long_name=H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10
profile=High
codec_type=video
codec_tag_string=avc1
….
[/STREAM]
[STREAM]
index=1
codec_name=mp2
codec_long_name=MP2 (MPEG audio layer 2)
profile=unknown
codec_type=audio
codec_tag_string=mp4a
codec_tag=0x6134706d
sample_fmt=fltp
sample_rate=44100
channels=2
channel_layout=stereo
…
[/STREAM]

 

The audio track is #2 (counted from 1), let’s add the language tag ‘en’:

mp4box -lang 2=en test_real_clock.mp4

Verification:

ffprobe -hide_banner -i test_real_clock.mp4 -show_streams

…

[STREAM]
index=1
codec_name=mp2
codec_long_name=MP2 (MPEG audio layer 2)
profile=unknown
codec_type=audio
codec_tag_string=mp4a
codec_tag=0x6134706d
sample_fmt=fltp
sample_rate=44100
channels=2
channel_layout=stereo

…
TAG:language=eng
TAG:handler_name=SoundHandler
TAG:vendor_id=[0][0][0][0]
[/STREAM]

Extract Elementary Stream

From ‘mp4box -info  mp4-file‘ you know that the track #1 is h264 video then the following command extract the elementary h264 stream from the mp4-file:

MP4Box -raw 1 input.mp4 -out es.h264

Extracting MPEG-4 AVC-H264 stream to h264

Note: if es.h264 exists mp4box does not overwrite it and does not extract data, although a corresponding print is issued. In such case i recommend to use ‘-new’ instead of ‘-out’, you get the error: “Error – 2 input names specified, please check usage”