Method and device for preprocessing streams of encoded data to facilitate decoding streams back-to back

• US 5,838,678 A
• Filed: 07/24/1996
• Issued: 11/17/1998
• Est. Priority Date: 07/24/1996
• Status: Expired due to Fees

• Current Assignee: Digital Video Systems Incorporated
• Original Assignee: Digital Video Systems Incorporated
• Inventors: Davis, Joseph W., Hayes, Shawn M.
• Primary Examiner(s): TON, DANG T
• Application Number:
  US08/686,629
• Time in Patent Office: 846 Days
• Field of Search: 370/389, 370/493, 370/496, 370/498, 370/535, 370/536, 370/537, 370/538, 370/539, 370/540, 370/541, 370/542, 370/543, 370/474, 370/476, 370/477, 348/461, 348/462, 348/466, 348/467, 348/390, 348/395, 348/396, 348/394, 348/404, 348/409, 348/12, 348/13, 348/19, 348/607, 348/391, 348/400, 348/412, 348/384
• US Class Current: 370/389
• CPC Class Codes: H04N 21/2335 involving reformatting operations of audio signals, e.g. by converting from one coding standard to another details of audio signal transcoding G10L19/173
• CPC Class Codes: H04N 21/234381 by altering the temporal resolution, e.g. decreasing the frame rate by frame skipping
• CPC Class Codes: H04N 21/23608 Remultiplexing multiplex streams, e.g. involving modifying time stamps or remapping the packet identifiers

Abstract

A method and device for preprocessing streams of encoded data (e.g. compressed in accordance with an MPEG standard) to permit a decoder to decode the streams, back-to-back (i.e., one stream immediately following another), without being reset and without producing video artifacts. The present invention includes verifying that a multiplexed stream complies with an encoding standard and preprocessing packets of the packetized encoded video sequence such that no video artifacts are produced when a video decoder decodes an adjacent encoded video sequence.

Claims

1. A method for preprocessing multiplexed streams of packets of packetized, encoded, audio and video sequences such that an audio decoder can decode the encoded audio sequence to produce an audio sequence and a video decoder can, without being reset, decode the encoded video sequence to produce a video sequence, the method including the steps of:

a) verifying that a multiplexed stream complies with an encoding standard for audio and video data;

b) preprocessing packets of the packetized, encoded, video sequence such that no video artifacts are produced when the video decoder decodes another encoded video sequence adjacent to the encoded video sequence; and

c) preprocessing the packets of packetized, encoded, audio data sequence such that a start time of the audio sequence is within a first predetermined time of a start time of the video sequence and a temporal length of the audio sequence is within a second predetermined time of a temporal length of the video sequence.

3. The method of claim 1 wherein the step of preprocessing the packets of the packetized, encoded, audio sequence includes sub-steps of:

i) removing any partial audio frames;

ii) adjusting the number of audio frames, if necessary, such that the start times of the audio and video sequences are within the first predetermined time; and

iii) adjusting the number of audio frames, if necessary, such that the temporal lengths of the audio and video sequences are within the second predetermined time.

2. A method for preprocessing multiplexed streams of packets of packetized, encoded, audio and video sequences such that an audio decoder can decode the encoded audio sequence to produce an audio sequence and a video decoder can, without being reset, decode the encoded video sequence to produce a video sequence, the method including the steps of:

a) verifying that a multiplexed stream complies with an encoding standard for audio and video data; and

b) preprocessing packets of the packetized, encoded, video sequence such that no video artifacts are produced when the video decoder decodes another encoded video sequence adjacent to the encoded video sequence, wherein the step of preprocessing the packets of packetized, encoded, video sequence includes the sub-steps of;

i) deleting any video frames that cannot be decoded if video frames of the video sequence are not temporally correct; and

ii) deleting any video frames following a code indicating an end of the encoded video sequence.

4. A method for preprocessing files including multiplexed streams of packets of packetized, compressed, audio and video sequences such that a video decoder can, without being reset, decode packets of the packetized, compressed, video sequence immediately following packets of another packetized, compressed, video sequence, the method comprising steps of:

a) verifying that a file complies with a compression standard;

b) adding a private data packet to the file;

c) remapping packet identifiers of the packets of packetized, compressed, audio and video sequences;

d) determining whether the packets of the packetized, compressed, video sequence pass a first set of tests; and

e) deleting selected ones of frames of the compressed video sequence if the packets of the packetized, compressed, video sequence were not determined to have passed the first set of tests in step (d).

5. The method of claim 4 further comprising steps of:

f) determining whether packets of the packetized, compressed, audio sequence, corresponding to the compressed video sequence, pass a second set of tests; and

g) adjusting a number of frames the compressed audio sequence if the packets of the packetized, compressed, audio sequence were not determined to have passed the second set of tests in step (f).

6. The method of claim 4 wherein the compressed audio and video sequences have been compressed in accordance with an MPEG, Motion Pictures Expert Group, standard, and

wherein the step of verifying includes at least one sub-step selected from a group consisting of:

i) determining whether the file has a predetermined number of sync bytes located 188 bytes apart;

ii) determining whether a first packet of a multiplexed stream has a program clock reference, PCR, discontinuity flag set;

iii) determining whether presentation time stamps, PTS and DTS, fields exist before the compressed video sequence;

iv) determining whether a PTS field exists before the compressed audio sequence;

v) determining whether the compressed video sequence begins with a sequence header;

vi) determining whether picture data of the video sequence starts with an I-Picture frame; and

vii) determining whether a first group-of-pictures of the compressed video sequence is closed.

7. The method of claim 4 wherein, in the step of adding a private data packet to the file, the private data packet contains at least one of (a) a source ID field, (b) a bit rate field, (c) a video frame count, and (d) a version field.

8. The method of claim 4 wherein, in the step of adding a private data packet to the file, the private data packet contains a bit rate field and a video frame count.

9. The method of claim 4 wherein the step of determining whether the packets of the packetized, compressed, video sequence pass the first set of tests includes sub-steps of:

i) determining whether frames of compressed video sequence are temporally correct; and

ii) determining whether any frames of the video sequence follow a code which indicates an end of the compressed video sequence.

10. The method of claim 9 wherein the step of deleting selected ones of frames of the compressed video sequence if the packets of packetized, compressed, video sequence were not determined to have passed the first set of tests in step (d) includes sub-steps of:

i) deleting frames of video data from the compressed video sequence until temporal correctness is restored; and

ii) deleting frames of video data from the compressed video sequence following the code.

11. The method of claim 5 wherein the step of determining whether the packets of the packetized, compressed, audio sequence pass the second set of tests includes sub-steps of:

i) determining whether any partial audio frames exist in the compressed audio sequence;

ii) determining whether the audio and video sequences start within a first predetermined time; and

ii) determining whether the audio sequence has a temporal length within a second predetermined time of a temporal length of the video sequence.

12. The method of claim 11 wherein the step of deleting selected ones of frames of the compressed audio sequence if the packets of the packetized, compressed, audio sequence were not determined to have passed the second set of tests in step (f) includes sub-steps of:

i) removing any partial audio frames in the compressed audio sequence;

ii) adjusting the number of audio frames in the compressed audio sequence until the start time of the audio sequence is within the first predetermined time of the start time of the video sequence; and

iii) adjusting the number of audio frames in the compressed audio sequence until the temporal length of the audio sequence is within the second predetermined time of the video sequence.

13. The method of claim 4 further comprising a step of updating the private data packet.

14. The method of claim 7 further comprising a step of updating the private data packet by modifying the video frame count to reflect any video frames deleted from the compressed video sequence.

15. A device for converting a transport stream file of MPEG encoded audio and video data to a preprocessed transport stream file of MPEG encoded audio and video data such that a video decoder can decode two consecutive preprocessed streams of the MPEG encoded video data without being reset, the device including:

a) a storage device for storing transport stream files and preprocessed transport stream files;

b) a program storage device for storing a sequence of instructions;

c) a controller for executing the sequence of instructions and controlling access to the transport stream files and the preprocessed transport stream files in the storage device;

d) a preprocessor for, under control of the controller, reading a transport stream file from the storage device, converting the read transport stream file into a preprocessed transport stream file, and writing the preprocessed transport stream file to the storage device;

e) a buffer memory for reading a preprocessed transport stream file from the storage device under control of the controller and buffering the read preprocessed transport stream file; and

f) an input/output interface for formatting the preprocessed transport stream buffered in the buffer memory for output to the decoder.

16. The device of claim 15 wherein the input/output interface is a SCSI-2 fast and wide interface.

17. The device of claim 15 further comprising a communications network arranged between the input/output interface of the device and the decoder.