Detecting markers in an encoded video signal

A computer program product comprising code stored on a computer:

• readable storage memory and configured, when run on a computer having multiple parallel processing units, to implement a video decoding method comprising:
• receiving a stream of data elements, some of which contain encoded video content, the stream comprising marker sequences, each marker sequence comprising a marker which does not contain the encoded video content, wherein a known pattern of data elements occurs in each marker sequence;
• supplying a respective part of the stream to each parallel processing unit;
• each parallel processing unit processing the respective part of the stream, whereby multiple parts of the stream are processed in parallel, by:
• forming a plurality of subsequences of the respective part of the stream;
• mapping each element of each subsequence to one of a first data element value or a second data element value to form mapped elements, the first data element value selected in the mapping to cause a saturation when summed with itself a number of times corresponding to a number of data elements in the known pattern;
• summing, for each subsequence, the mapped elements of the subsequence to form a plurality of sums, the second data element value selected in the mapping to prevent a saturation when the second data element value is included in the summing;
• and
• detecting that the respective part of the stream includes the known pattern of data elements, thereby identifying the markers, the detecting based on at least one of the plurality of sums indicating a saturation;
• separating the encoded video content from the identified markers;
• decoding the separated video content;
• and
• outputting the decoded video content on a display.

A computer having multiple parallel processing units and comprising:

• a display;
• an input configured to receive a stream of data elements, some of which contain encoded video content, the stream comprising marker sequences, each marker sequence comprising a marker which does not contain the encoded video content, wherein a known pattern of data elements occurs in each marker sequence;
• a content separator module implemented at least partially in hardware of the computer to supply a respective part of the stream to each parallel processing unit, each parallel processing unit configured to process the respective part of the stream, whereby multiple parts of the stream are processed in parallel by:
• forming a plurality of subsequences of the respective part of the stream;
• mapping each element of each subsequence to one of a first data element value or a second data element value to form mapped elements, the first data element value selected in the mapping to cause a saturation when summed with itself a number of times corresponding to a number of data elements in the known pattern;
• summing, for each subsequence, the mapped elements of the subsequence to form a plurality of sums, the second data element value selected in the mapping to prevent a saturation when the second data element value is included in the summing;
• and
• detecting that the respective part of the stream includes the known pattern of data elements, thereby identifying the markers, the detecting based on at least one of the plurality of sums indicating a saturation, the content separator configured to separate the encoded video content from the identified markers;
• and
• a video decoder configured to decode the separated video content and output the decoded video content on the display.