Video and audio information processing

Claim 2

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from said image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data including color distribution data associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images, and including sub shot segmentation data derived from said color distribution data; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus.

Claim 3

The apparatus according to claim 1 , in which said image feature vector data includes face recognition data.

Claim 4

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data including color distribution data associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images, said image property data including activity measure data derived from a variance of said color distribution data and indicative of a change of said image content or said audio content between said video images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus.

Claim 5

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data including color distribution data associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images, said image property data including activity measure data derived from a variance of said color distribution data and indicative of a change of said image content or said audio content between said video images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus.

Claim 6

The apparatus according to claim 3 , in which said image feature vector data includes face recognition data.

Claim 7

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from said image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data including color distribution data associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images, said image property data includes a representative key frame derived from said color distribution data and indicative of a predominant overall content of said video images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus.

Claim 8

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from said image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data including color distribution data associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images, said image property data includes a representative key frame derived from said color distribution data and indicative of a predominant overall content of said video images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus.

Claim 9

The apparatus according to claim 5 , in which said image vector data includes face recognition data.

Claim 10

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from said image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data being associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images or groups of images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus, in which: said camera-recorder apparatus is operable to capture an audio signal associated with said video images; said feature extraction unit is operable to derive audio feature vector data identifying speech content for portions of said audio signal associated with at least one of said video images; and said image property data includes interview detection data indicative of an interview sequence of said video images, said video images of said interview sequence including identified facial images co-occurring with respect to said audio signal that is associated with said video images of said interview sequence comprising speech.

Claim 11

A camera-recorder apparatus comprising: an image capture device operable to capture a plurality of video images; a storage medium by which said video images are stored for later retrieval; a feature extraction unit operable to derive image feature vector data from said image content of at least one of said video images substantially in real time at said capture of said video images, said image feature vector data being associated with respective images; a metadata extraction unit operable to derive image property data from said image feature vector data substantially in real time at said capture of said video images, said image property data being associated with said respective images or groups of images; and a data path by which said camera-recorder apparatus is operable to transfer said derived image property data to an external data processing apparatus, in which: said camera-recorder apparatus is operable to capture an audio signal associated with said video images; said feature extraction unit is operable to derive audio feature vector data identifying speech content for portions of said audio signal associated with at least one of said video images; and said image property data includes interview detection data indicative of an interview sequence of said video images, said video images of said interview sequence including identified facial images co-occurring with respect to said audio signal that is associated with said video images of said interview sequence comprising speech.

Claim 12

The apparatus according to claim 7 , in which said image feature vector data includes at least color distribution data and/or face recognition data.

Claim 13

The apparatus according to claim 7 , in which said audio feature vector data comprises speech detection data.

Claim 14

The apparatus according to claim 9 , in which: said audio feature vector data comprises speech detection data; and said metadata extraction unit is operable to derive said interview detection data from said face recognition data and said speech detection data.