Apparatus and Method for Storing and Reading a File having
a Media Data Container and a Metadata Container
Specification
The invention relates to media storage, transmission,
reception and playback, in particular to media storage in
or playback from a file having a media data container and a
metadata container, as e.g. a file based on the ISO
(International Organization for Standardization) base media
file format.
Various electronic devices are enabled to receive and
present media data streams. Such media data streams can
e.g. be received from a digital video broadcasting network
that broadcasts media streams in accordance with e.g. the
DVB-H Standard (Digital Video Broadcasting - Handhelds) or
the DVB-T Standard (Digital Video Broadcasting
Terrestrial) .
DVB-T uses a self-contained MPEG-2 (MPEG = Moving Pictures
Expert Group) transport stream containing elementary MPEG-2
video and audio streams according to the international
standard ISO/IEC 13818 (IEC = International
Electrotechnical Commission). The MPEG-2 transport stream
is a multiplex used in many of today's broadcast systems.
It is a stream multiplex of one or more media programs,
each containing typically audio and video but also other
data. MPEG-2 transport streams share a common clock per
program and use time-stamped media samples (Access Units,
AUs) in all media streams within a program. This enables
synchronization of sender and receiver clocks and lip
synchronization of audio and video streams.
For DVB-H, elementary audio and video streams are
encapsulated in RTP (Real-Time Transport Protocol), UDP
(User Datagram Protocol), IP (Internet Protocol), and MPE
(Multi-Protocol Encapsulation) for IP data casting. RTP is
used for effective real-time delivery of multi-media data
over IP networks. Multiplexing is typically done by
associating different network ports to each distinct media
stream, e.g. one network port for video and another one for
audio.
A streaming service is defined as a set of synchronized
media streams delivered in a time-constraint or
unconstraint manner for immediate consumption during
reception. Each streaming session may comprise audio, video
and/or real-time media data like timed text. A user
receiving media data for a movie by means of a mobile
television, for instance, can watch the movie and/or record
it to a file. Commonly, for this purpose the received data
packets of the received media stream are de-packetized in
order to store raw media data to the file. That is,
received RTP packets or MPEG-2 packets are first de-
packetized to obtain their payload in form of media data
samples, such as compressed video or audio frames. Then,
after de-packetizing, obtained media data samples are
replayed or stored to the file. The obtained media samples
are commonly compressed by formats like the H.264/AVC (AVC
= Advanced Video Coding) video format and/or the MPEG-4 HE-
AACv2 (HE-AACv2 = High-Efficiency Advanced Audio Coding
version 2) audio format. When media data samples having
such video and/or audio formats are to be stored, they may
be stored in a so-called 3GP file format, also known as
3GPP (3rd Generation Partnership Project) file format, or
in an MP4 (MPEG-4) file format. Both 3GP and MP4 are
derived from the ISO base media file format, which is
specified in the ISO/IEC international standard 14496-
12:2005 "Information technology-coding of audio-visual
objects - part 12: ISO base media file format". A file of
this format comprises media data and metadata. For such a
file to be operable, both of these data must be present.
The media data is stored in a media data container (mdat)
related to the file and the metadata is stored in a
metadata container (moov) of the file. Conventionally, the
media data container comprises actual media samples. I.e.,
it may comprise e.g. interleaved, time-ordered video and/or
audio frames. Thereby, each media has its own metadata
track (trak) in the metadata container moov that describes
the media content properties. Additional containers (also
called boxes) in the metadata container moov may comprise
information about file properties, file content, etc.
Recently, so-called reception hint tracks for files based
on the ISO base media file format have been defined by
international standardization groups. Those reception hint
tracks may be used to store multiplexed and/or packetized
streams like e.g. a received MPEG-2 transport stream or RTP
packets. Reception hint tracks may be used for a client
side storage and playback of received data packets. Which
shall also be denoted as data samples in the seguel of this
specification. Thereby, received MPEG-2 TS or RTP packets
of one stream are directly stored in reception hint tracks
as e.g. pre-computed samples or constructors. I.e., in the
case of reception hint tracks, the data packets are stored
as samples in the media data container of the file based on
the ISO base media file format. Playback from reception
hint tracks may be done by emulating the normal stream
reception and reading the stored data packets from the
reception hint track as they were received over IP.
The ISO/IEC international standard 14496-12:2005
"Information technology-coding of audio-visual objects -
part 12: ISO base media file format" defines a sample
grouping as an assignment of each sample in a track to be a
member of one sample group, based on a grouping criterion.
As there may be more than one sample grouping for the
samples in track, each sample grouping has a type field to
indicate the type of grouping.
Sample groups are defined in two steps. First, a type of
the grouping is defined in a sample group description box
(sgpd) . In a second step, this description is assigned to
samples in a sample-to-group box (sbgp). The sample groups
mechanism is extensible and is currently used for AVC- and
SVC-specific extensions and proprietary extensions.
A non-exhaustive description of the syntax is given below:

A simplified version of the SampleGroupDescriptionBox is
given here. In the ISO file format specialized versions
depending on the handler type exist.

In one instance of the box multiple groups can be defined
and every sample must be member of one group. The syntax of
the SampleToGroup box is provided.

The following abstract example shall illustrate how sample
groups work:
Let us assume that the "color" of each sample has to be
described. For a complete set of samples all samples with
the same color are grouped together.
First, it has to be specified which colors can occur. For
each color, a "SampleGroupDescriptionEntry" is defined. A
value for the grouping_type "color" is defined and all
color description entries are stored in the
SampleGroupDescriptionBox for the grouping_type color.
Second, the sample-to-group box for the "color"
grouping_type describes which sample has which color. This
is done in differential way: every list entry describes how
many consecutive samples have the same color. This allows a
very compact storage for a rare change of colors, e.g.
first a high number of samples have color one, then a
number of samples have color 2 and so on.
For three colors and a file of 50 samples the tables based
on the above described syntax could look like this:

As described above, sample groups are well suited to
classify samples into different categories, but they are
not well suited when events related to or properties of
individual samples need to be described in the file. The
main reason for that is that sample groups always describe
a complete set of samples, and samples that do not belong
to a group entry must be member of a "does not belong to
any group"-group entry. Another reason is a slow look-up of
the sample group a sample belongs to.
An event or property shall be understood as an index for a
single sample or a relatively small number of samples. The
event or property occurs on an indexed sample, but may
influence an arbitrary number of following samples, e.g.
random-access-points can be treated as events.
An example for events compared to the above example is
"color-change". If not the color itself, but the change
from on color to another has to be indexed, sample groups
are not very well suited, because the "sample group" based
index has to include also the unwanted information "no
color change". Especially in the case of frequent changes,
this may lead to an inefficient index table. Parsing for
the events near the end of file tends to be a complex
operation, because . all sample counts (also that of the
"non-event" group samples) have to be summed up.
For trick-play modes (e.g. fast-forward, seeking into the
file, etc.) the closest random-access-point to the desired
entry point needs to be identified efficiently. Therefore a
table of samples this event applies to must be examined for
the right entry-point. Random-access-points can exist at
multiple levels, so, e.g., first the video decoder
configuration is needed in the file and then the closest I-
frame of the video track and above of that the multiplex-
level entry-point (e.g. the PAT in case of MPEG-2 TS).
An additional problem is that sample groups do not allow
the association of a sample to multiple group descriptions.
This complicates stacking of events and will not give a
compact representation, if sample groups are used for
solving this indexing issue.
Hence, it is an object of the present invention to provide
a concept for efficiently associating events or properties
related to a data sample in a file having a media data
container and a metadata container, allowing for a
efficient look-up of said sample.
This object is solved by an apparatus for recording a file
according to claim 1, a method for recording a file
according to claim 17, an apparatus for reading a file
according to claim 19 and a method for reading a file
according to claim 26.
For the solution of above-mentioned object, embodiments of
the present invention also provide computer-programs for
carrying out the inventive methods.
The present invention is based on the finding that an event
or a property related to a specific sample or to a certain
number of subsequent samples can be provided by storing
property information together with the related sample
number in the metadata container of a file based on the ISO
base media file format. In a specific embodiment the
property information relates to errors of or related to a
certain sample or a sequence of samples. For the purpose of
storing error-related metadata, dedicated containers or
boxes are provided in the metadata container ("moov") of
the file. Thereby, error information to be stored is
defined in two steps. First, a type of error is defined in
a sample-property-description box ("spdb"). In a second
step, the error type description is assigned to specific
samples in a sample-to-property box ("stpb").
According to a preferred embodiment of the present
invention the property- or error-related metadata boxes
spdb, stpb are comprised by a sample-table box ("stbl")
comprising time and data indexing of the samples in a
track. Using the tables comprised by the sample-table box
stbl, it is possible to locate samples in time, determine
their type, their size, container (in the media data
portion of the file) and offset into that container.
With embodiments of the present invention it is possible to
efficiently determine erroneous samples or samples in the
neighborhood of erroneous samples. For example, an
erroneous sample might be a corrupted sample or a missing
sample that was either not stored in the media data
container or lost during a previous transmission of data
samples from a transmitter to a receiver before saving it
to the file.
According to an embodiment of the present invention the
samples are data packets, such as RTP or MPEG-2 transport
stream data packets, received during a streaming session or
already stored in the media data container of the file
based on the ISO base media file format as in the case of
reception hint tracks.
According to an embodiment of the present invention the
error information stored in the metadata container (moov)
comprises a qualitative error information, such as an error
type, and the associated sample number of the erroneous
data sample. Additionally, the error information may
further comprise quantitative error information, such as
e.g. a detailed description of an error.
In order to generate the file, embodiments of the present
invention provide an apparatus for outputting the file
having a media data container and a metadata container, the
apparatus comprising an error information provider for
providing an error information related to a data sample,
and a recorder for storing the error information together
with a sample number related to the data sample in the
metadata container.
Further, embodiments of the present invention provide an
apparatus for reading a file with a media data container
having stored data samples, and with the metadata container
having stored error information related to the stored data
samples, the apparatus comprising a parser for parsing the
metadata container in order to find error information
related to a data sample to be processed, and a processor
for providing an error-specific measure or action in case
the related error information indicates that the data
sample to be processed is erroneous.
For example, the error-specific measure may be an error
concealment or an error indication measure. Error
concealment may be performed in order to hide an erroneous
data sample or data packets from a user e.g. during
playback of the stored data samples or packets. For
example, when replaying audio, a lost or corrupted audio
frame may be replaced by specific concealment methods known
in the art. The same holds for lost or corrupted video
frames.
Further, the stored error information may be used to
unambiguously identify all erroneously received data
samples after the complete transmission of data samples is
received and stored to a file. Embodiments of the present
invention provide an apparatus for reading and parsing the
metadata container in order to identify erroneous data
samples stored in the media data container and request and
receive new copies of these erroneous data samples from a
data server, e.g. a streaming server, in a non-realtime
operation. The erroneous data samples are replaced in the
file with the newly received error-free data samples to
convert the stored file containing errors to an error-free
file.
Embodiments of the present invention offer a compact
representation of errors. Sample numbers may be used for a
fast table look-up. This may be especially useful for long
files, like e.g. recordings of complete movies. By directly
relating error information to sample numbers a depacketizer
or decoder may be "warned" or informed before trying to
process an erroneous data packet or data sample, such that
appropriate countermeasures can be envisaged.
Further objects and features of the present invention will
become apparent from the following detailed description
considered in conjunction with the accompanying drawings,
in which:
Fig. la shows a schematic block diagram of an apparatus
for outputting a file having a media data
container and a metadata container according to
an embodiment of the present invention;
Fig. lb shows a schematic block diagram of an apparatus
for outputting a file having a media data
container and a metadata container according to a
further embodiment of the present invention;
Fig. 1c shows a schematic block diagram of an apparatus
for outputting a file having a media data
container and a metadata container according to
yet a further embodiment of the present
invention;
Fig. 2 shows a schematic structure of a sample table box
according to an embodiment of the present
invention;
Fig. 3 shows a flow chart of a method for outputting
said file according to an embodiment of the
present invention; and
Fig. 4 shows a schematic block diagram of an apparatus
for reading a file having a media data container
and a metadata container according to the
embodiment of the present invention.
The following description sets forth specific details, such
as particular embodiments, procedures, techniques, etc. for
purposes of explanation and not limitation. But it will be
appreciated by one skilled in the art that other
embodiments may be employed apart from these specific
details. For example, although the following description is
facilitated using non-limiting example applications to
different embodiments, the technology may be employed to
any type of systems. In some instances, detailed
descriptions of well known methods, interfaces, circuits,
and device are omitted so as not obscure the description
with unnecessary detail. Moreover, individual blocks are
shown in some of the figures. Those skilled in the art will
appreciate that the functions of those blocks may be
implemented using individual hardware circuits, using
software programs and data, in conjunction with a suitably
programmed digital microprocessor or general purpose
computer, using application specific integrated circuitry
(ASIC), and/or using one or more digital signal processors
(DSPs).
Fig. la shows an exemplary block diagram of an apparatus 10
for outputting a file 12 having a media data container 14
and a metadata container 16.
The apparatus 10 comprises an error information provider 18
for providing an error information 19 related to a data
sample 20 input to the error information provider 18.
Further, the apparatus 10 comprises a recorder 22 for
storing the error information 19 together with a sample
number 21 related to the data sample 20 in the metadata
container 14 of the file 12.
According to an embodiment of the present invention the
error information provider 18 is adapted to analyze a
sequence of data samples 20 in order to provide an error
information 19 related to missing data samples in the
analyzed sequence of data samples. This may be useful for
identifying samples that were, for example, lost during a
streaming session. In case a missing data sample is
detected, the recorder 22 may store error information 19
indicating the missing data sample together with a sample
number 21 of an existing data sample next to or in the
neighborhood of the missing data sample or packet. In other
words, since the recorder 22 cannot store a sample number
of a not existing because lost data sample, it may
associate the error information related to the lost sample
to a neighboring data sample that was not lost.
Illustrative examples will be given further below.
Additionally or alternatively, the error information
provider 18 may be adapted to detect whether at least a
part of the data sample 20 contains corrupted information.
For example, due to certain effects, e.g. like channel
fading or additive noise, during the transmission from data
samples from a sender to a receiver, information carried by
the data samples may be tampered or corrupted, such that a
received version of a data sample does not correspond to a
transmitted version of the data sample anymore. In this
case, error correction or error concealment methods may
e.g. be applied at the receiving end.
Fig. lb shows an apparatus 10 according to a further
embodiment of the present invention.
In the embodiment shown in Fig. lb data samples 20-1, 20-2,
20-3 are already stored in the media data container 16
(mdat) of the file 12 based on the ISO base media file
format. Typically, samples 20-1, 20-2, 20-3 within the
media data box 16 are grouped into so-called chunks. Chunks
can be of different sizes, and the samples within a chunk
can have different sizes. In the case of Fig. lb the error
information provider 18 is adapted to parse or analyze the
stored data samples 20-1, 20-2, 20-3 in the media data
container 16 in order to detect missing, corrupted or
generally erroneous data samples. In case the data samples
20-1, 20-2, 20-3 are data packets, such as e.g. RTP or
MPEG-2 transport stream data packets, the data packets
usually comprise sequence numbers indicating an order of
transmission. The error information provider 18 may check
the sequence numbers of the stored data samples 20-1, 20-2,
20-3 and thus detect missing sequence numbers.
Additionally, the stored sequence or chunks of data samples
20-1, 20-2, 20-3 may be parsed in order to find corrupted
samples.
A further embodiment of the apparatus 10 is shown in
principle in Fig. 1c.
This embodiment comprises a receiver 24 for receiving
streamed data samples 20-1, 20-2, 20-3, which may be raw
media data samples or data packets comprising packetized
media data samples. The output of the receiver 24 is
coupled to the input of the error information provider 18,
such that it may check the received data samples 20-1, 20-
2, 20-3 in order to detect a missing or corrupted data
sample and to provide according qualitative and,
optionally, quantitative error information. In the example
given in Fig. 1c the recorder 22 is adapted to store the
received data samples 20 in chunks of the media data
container 16 of the file ,12 and to associate a sample
number 21 to each of the stored data samples.
According to embodiments of the present invention the
provided error information 19 is stored in a sub-box of a
sample table box (stbl) comprised by the metadata container
14 (moov), the sample table box (stbl) allowing an indexing
from timing of the data samples 20-1, 20-2, 20-3 to their
associated sample numbers 2-1-1, 21-2, 21-3 in the chunks.
This shall be illustrated in further detail referring to
Fig. 2.
Fig. 2 shows a sample table box stbl 50 of a file based on
the ISO base media file format comprising - besides
conventional media data referencing sub-boxes, such as e.g.
a sample description box (stsd), a sample size box (stsz),
a sample to chunk box (stsc) and a chunk offset box (stco)
- inventive sub-boxes 52 and 54. The sample description box
(stsd) is required because it contains a data reference
index field which indicates, which data reference box is
used to retrieve the samples in the media data container
(mdat). Without the sample description it would not be
possible to determine where the samples are stored. A synch
sample table (stss) is optional. If the synch sample table
(stss) is not present, all samples are synch samples. In
addition to the conventional sub-boxes known to a person
skilled in the art, the sample table box stbl 50 comprises
two additional sub-boxes 52 and 54 related to the error
information 19 provided by the error information provider
18. In the following, sub-box 52 shall be exemplarily
denoted as SamplePropertyDescriptionBox spdb, wherein sub-
box 54 shall be exemplarily denoted as SampleToPropertyBox
stpb. According to embodiments of the present invention, a
property of a sample is the fact that the sample relates to
an error or that the sample itself is erroneous.
The SampleToPropertyBox stpb 54 maps samples to their
properties, i.e. errors according to embodiments.
Properties may be stacked, i.e. multiple properties of the
same property type (error) may apply to one sample. A
length of a value of a specific property (property_length)
is unspecified and depends on the property. A file reader
that does not understand a property-type may discard the
entire box stpb 54.
The SamplePropertyDescriptionBox spdb 52 describes the
properties of one property_type. A sample property
description entry is not specifically defined and may be
proprietary. A reading apparatus may parse the box spdb 52
only if it understands property_type or will discard both
the SampleToPropertyBox stpb 54 and the
SamplePropertyDescriptionBox spdb 52 of this property__type.
An examplary syntax for sample properties may look like
this:

According to embodiments errors may be indexed, i.e.
transmission errors may be marked in a reception hint
track. For this reason two different error classes may be
defined:
a) lost packets
b) corrupted packets
Hence, according to the exemplary nomenclature above, a
SamplePropertyDescriptionEntry may be defined for each of
these according to

The SamplePropertyDescriptionEntries may be defined as:

unsigned int(32) desc_type = 'lost';
unsigned int(16) property_length = 32; // value contains
number of consecutively lost packets before
unsigned int(8) verbose_description = "lost transmission
packets"; }

Here, LostPacketEntry corresponds to first qualitative
error information and is used to initialize the run of the
first property_desc_index in the SampleToPropertyBox stpb
54. CorruptedPacketEntry corresponds to second qualitative
error information and initializes the run with the second
property_desc_index.
If we have for example a transmission with packets [1, n],
with n = 1000, which was affected by transmission errors,
say packets 310 to 367 were lost and packets 34 and 177
were corrupted, the entries of the SampleToPropertyBox may
be:

If each received packet yields a sample and packets 310 to
367 were lost, then samples 1 to 309 contain packets 1 to
309. Samples 310 to 942 contain packets 378 to 1000.
In the example above, for property_desc_index with value 1,
the next received packet after the loss contains the error
indication or information. The value of the SampleProperty,
i.e. the quantitative error information, is the number of
consecutively lost packets before, hence 58.
Corrupted packets were marked as such with the run for
property_desc_index = 2. Hence, sample 34 and 177 were
marked as corrupted. Since the value of property_length is
zero for this property_desc_index, no value is assigned to
it.
Turning now to Fig. 3, a method for outputting a file 12
based on the ISO base media file format shall be
summarized.
In a first step S1 an error information provider 18
analyzes a data sample or a data packet 20 in order to
determine an error information 19 related to the analyzed
sample 20. In case an error yielding an error information
is detected related to said sample, the error information
19 is stored, together with its associated sample number 21
of the erroneous data sample, in the metadata container 14
in step S2. Step S2 comprises storing a
SamplePropertyDescriptionBox 52 and a SampleToPropertyBox
54 in the sample table stbl 50 of the metadata container
moov 14 of the file 12.
If the data sample 20 is received, e.g. in case of a
reception of streamed data samples, each of received data
samples is associated thereto a sample number 21. Further,
each of received data samples may be stored in chunks of
the media data container 16 in an optional step S3.
It can be seen that step S2 and optional step S3 may be
essentially performed in parallel, i.e. the error
information 19 and the sample number 21 are stored in the
media data container 14 while the data sample is stored in
the media data container 16 at the same time.
As has been explained before, the samples may be data
packets, such as RTP or MPEG-2 transport stream data
packets, as it is the case when storing reception hint
tracks, as explained in the introductory portion of this
specification.
After the samples have been stored in the media data
container 16 and the error information has been stored in
the metadata container 14 of the file 12 based on the ISO
base media file format, the stored error information may be
used e.g. during a playback of the samples. For that
purpose, embodiments of the present invention provide an
apparatus 60 for reading the file 12 with the media data
container 16 and with the metadata container 14, as
exemplarily shown in Fig. 4.
The apparatus 60 comprises a parser 62 for parsing or
analysing the metadata container 14 in order to find error
information 19 related to a data sample 20 to be processed.
A processor 64 may provide an error-specific measure in
case the related error information 19 indicates that the
data sample 20 to be processed is erroneous.
To be more specific, the parser 62 may parse the
SamplePropertyDescriptionBox 52 and the SampleToPropertyBox
54 in the sample table stbl 50 of the metadata container
moov 14 of the file 12. Thereby it may look for sample
numbers 21 and their associated qualitative error
information 19, e.g. their property_desc_index when
referring to the nomenclature used above. Additionally the
parser 62 may extract quantitative error information, e.g.
how many data packets were lost before or after the packet
having the associated sample number.
According to embodiments of the present invention, the
processor 64 is adapted to perform an error concealment
measure as the error-specific measure in response to the
detected error-information. For example, if the data sample
to be processed is corrupted or lost, the processor 64 may
synthesize the sample by performing some sort of spectral
interpolation between spectral values of neighbouring,
existing data samples. This may be done, for example, if
the data sample relates to an audio or video frame.
In case of a corrupted sample or data packet, the processor
64 may also be adapted to initiate some sort of error
correction, for example by employing channel decoding
means.
In a further embodiment, the processor 64 may also indicate
the error to a downstream device or to a user interface by
reporting the qualitative and quantitative error
information mentioned before.
Further, the stored error information 19 in the metadata
container 14 may be used to unambiguously identify
erroneously received data samples after a complete
transmission of data samples has been received and stored
to the file 12. For that reason the parser 62 may parse the
stored error information 19 in the metadata container 14 in
order to identify erroneous data samples stored in the
media data container 16. The processor 64 may request and
receive new copies or versions of the identified erroneous
data samples from a content provider's server, e.g. a
streaming server, in a non-realtime operation. I.e. no
realtime protocol is required for the retransmission of the
identified data samples, which may be single erroneous
video frames, for example. The erroneous data samples may
then be replaced in the file 12 by the requested and newly
received error-free data samples in order to convert the
stored file 12 containing errors to an error-free file.
The apparatus 60 may get an external request that creates
responses in form of error information for, e.g.
get response consisting of the error information that
applies to a particular requested sample,
get response with the closest sample number a
particular requested error (e.g. lost samples) applies
to.
The described events or sample properties in form of error
information are useful for e.g. indexes of reception hint
tracks.
Embodiments of the present invention offer the advantage
that errors related to samples can be described in an
extensible way. Multiple corresponding errors can be
grouped to an error-type. The quantitative error
information is not limited as it is the case, for example,
by the number of SampleGroupDescriptionEntries.
Further, a simple and efficient way is provided to
signalize error occurrence locations in a file. This may be
useful for error correction, error concealment and/or error
indication methods when replaying the data content stored
in the media data container of the file.
Depending on the circumstances, the inventive methods may
be implemented in hardware or software. The implementation
may be done on a digital storage medium, particularly a
disc, CD or DVD with electronically readable control
signals, which may cooperate with a programmable computer
system such that the method is executed. In general, the
invention thus also consists in a computer program product
with a program code stored on a machine-readable carrier
for performing the inventive method when the computer
program product runs on a computer. In other words, the
invention may thus be realized as a computer program with a
program code for performing the method when the computer
program runs on a computer.
While this invention has been described in terms of several
preferred embodiments, there are alterations, permutations
and equivalents which fall within the scope of this
invention. It should also be noted that there are many
alternative ways of implementing the methods and
compositions of the present invention. It is therefore
intended that the following appended claims be interpreted
as including all such alterations, permutations and
equivalents as falling within the true spirit and scope of
the present invention.
We claim:
1. Apparatus (10) for outputting a file (12) based on the
ISO base media file format having a media data
container (16) and a metadata container (14), the
apparatus comprising:
an error information provider (18) adapted to analyze
a sequence of data samples in order to provide an
error information (19) related to a missing or
corrupted data sample in the sequence of data samples;
a recorder (22) for storing the error information
together with a sample number (21) related to the
missing or corrupted data sample in the metadata
container (14) of the file (12) based on the ISO base
media file format.
2. The apparatus according to claim 1, wherein the
recorder (22) is adapted to store, in case of a
detected missing data sample in the sequence of data
samples, error information (19) indicating the missing
data sample together with a sample number (21) of an
existing data sample next to the missing data sample.
3. The apparatus according to claim 1, wherein the error
information provider (18) is adapted to detect,
whether at least a part of a data sample (20) contains
corrupted information.
4. The apparatus according to one of the preceding
claims, wherein data samples (20) are stored in the
media data container (16) of the file (12), and
wherein the error provider (18) is adapted to parse
the stored data samples in order to detect a missing
or a corrupted data sample.
5. The apparatus according to one of claims 1 to 3, the
apparatus comprising a receiver (24) for receiving
streamed data samples, and wherein the error provider
(18) is adapted to check the received data samples in
order to detect a missing or a corrupted data sample.
6. The apparatus according to claim 5, wherein the
recorder (22) is adapted to store the received data
samples in chunks of the media data container (16) of
the file (12) and to associate a sample number (21) to
each of the stored data samples.
7. The apparatus according to claim 6, wherein the
recorder (22) is adapted to store the provided error
information (19) in a sample table container (stbl)
allowing an indexing from timing of a stored data
sample to its associated sample number (21) in the
chunk.
8. The apparatus according to claim 7, wherein the
recorder is adapted to store a chunk offset table
(stco) indicating an index of each chunk into the file
(12) .
9. The apparatus according to one of the preceding
claims, wherein the error provider (18) is adapted to
associate the provided error information (19) to at
least one of a plurality of error types, each
indicating a different type of error.
10. The apparatus according to claim 9, wherein the
recorder (22) is adapted to store the at least one
error type in a related error type description
metadata container (52; spdb) comprised by a sample
table container (stbl) allowing an indexing from
timing of a data sample to its associated sample
number (21).
11. The apparatus according to one of the preceding
claims, wherein the recorder (22) is adapted to store
the provided error information (19) together with the
sample number (21) in an error property metadata
container (54; stpb) allowing an allocation between
the data sample and the provided error information
(19), wherein the error property metadata container
(54; stpb) is comprised by a sample table container
(stbl) allowing an indexing from timing of a data
sample to its associated sample number (21).
12. The apparatus according to one of the preceding
claims, wherein the provided error information (19)
comprises a qualitative and a quantitative error
information, and wherein the recorder is adapted to
store the quantitative error information together with
the qualitative error information and the sample
number (21), the quantitative error information
quantitatively characterizing the qualitative error
information.
13. The apparatus according to one of the preceding
claims, wherein a data sample (20) is a streamed data
packet comprising media data samples.
14. The apparatus according to claim 13, wherein the data
packet (20) is a streamed RTP packet, RTCP packet or
MPEG-2 transport stream packet.
15. Method for outputting a file (12) based on the ISO
base media file format having a media data container
(16) and a metadata container (14), the method
comprising:
analyzing a sequence of data samples in order to
provide an error information (19) related to a missing
or corrupted data sample in the sequence of data
samples;
storing (S2) the error information (19) together with
a sample number (21) related to the missing or
corrupted data sample in the metadata container of the
file (12) based on the ISO base media file format.
16. A computer program for carrying out the method
according to claim 15, when the computer program is
run on a computer or micro controller.
17. Apparatus (60) for reading a file (12) based on the
ISO base media file format with a media data container
(16) having stored data samples (20), and with a
metadata container (14) having stored error
information (19) related to the stored data samples,
the apparatus comprising:
a parser (62) for parsing the metadata container in
order to find error information related to a data
sample to be processed; and
a processor (64) for providing an error-specific
measure in case the related error information (19)
indicates that the data sample to be processed is not
available or corrupted.
18. The apparatus according to claim 17, wherein the
processor (64) is adapted to perform an error
concealment measure as the error-specific measure.
19. The apparatus according to claim 17, wherein the
processor (64) is adapted to perform an error
indication measure as the error-specific measure.
20. The apparatus according to claim 17, wherein the
processor (64) is adapted to request and newly receive
an error-free copy of a missing or corrupted data
sample, and wherein the processor (64) is adapted to
place the requested and newly received error-free data
sample in the file (12) in order to convert the stored
file comprising the corrupted data sample into an
error-free file.
21. The apparatus according to one of claims 17 to 20,
wherein the data samples are streamed data packet
comprising media data samples.
22. The apparatus according to claim 21, wherein the
stored data samples are RTP packets, RTCP packets or
MPEG-2 transport stream packets.
23. Method for reading a file (12) with a media data
container (16) based on the ISO base media file format
having stored data samples (20) , and with a metadata
container (14) having stored error information (19)
related to the stored data samples, the method
comprising:
parsing the metadata container (14) in order to find
error information (19) related to a data sample to be
processed; and
providing an error-specific measure in case the
related error information (19) indicates that the data
sample to be processed is missing or corrupted.
24. A computer program for carrying out the method
according to claim 23, when the computer program is
run on a computer or micro controller.

A concept is disclosed for outputting a file (12) having a
media data container (16) and a metadata container (14),
the concept comprising providing an error information (19)
related to a data sample (20) and storing the error
information together with a sample number (21) related to
the data sample in the metadata container (14).