Compressed video signal including information for independently coded regions
First Claim
1. For use with a predictive field or frame based encoded bitstream, each field or frame having multiple decodable subsets that can independently be recognized directly from the bitstream, an improvement comprising:
- a region identification signal stored on machine readable media;
said region identification signal identifying all decodable subsets that represent an independently encoded object appearing as a subset of a field or frame;
wherein said signal is adapted to permit extraction of data representing the independently encoded object by decoding the identified subsets directly from the bitstream;
at least one region map, corresponding to the region identification signal, the region map providing an indication for each and every separately decodable subset of the encoded field or frame representation of the image whether that separately decodable subset is required to be decoded in order to completely decode the subset of the image;
wherein each separately decodable subset is an image slice;
each region map includes multiple entries, including one for each slice for the encoded field or frame, where each entry indicates whether the associated slice includes image data corresponding to the independently encoded region associated with the region map; and
each region map provides an indication of those slices that must be decoded in order to completely decode the independently encoded region from a field or frame encoded representation of the associated image without decoding other image slices of the field or frame encoded representation.
2 Assignments
0 Petitions
Accused Products
Abstract
In a MPEG or other video system, a “region map” is preferably defined in header information for each frame which indicates for each independently coded region in the frame which “image slices” contain data for that region. Using a system of globally and locally defined regions (region groups), the map permits automatic identification of an object through multiple frames (notwithstanding movement) and for select decoding and extraction of that object only, without necessarily decoding all irrelevant image data. An edited region can be re-inserted into the original encoded bit stream in place of the extracted data using minimal computational resources. Using this signal format, objects or locations within a video sequence can be organized into region groups such that they can be tracked from frame to frame and associated with each other; each region group is independently coded, such that motion vectors and residuals of a dependent frame (such as a MPEG “P” or “B” frame) point only to a corresponding region in an anchor frame.
158 Citations
18 Claims
-
1. For use with a predictive field or frame based encoded bitstream, each field or frame having multiple decodable subsets that can independently be recognized directly from the bitstream, an improvement comprising:
-
a region identification signal stored on machine readable media; said region identification signal identifying all decodable subsets that represent an independently encoded object appearing as a subset of a field or frame; wherein said signal is adapted to permit extraction of data representing the independently encoded object by decoding the identified subsets directly from the bitstream; at least one region map, corresponding to the region identification signal, the region map providing an indication for each and every separately decodable subset of the encoded field or frame representation of the image whether that separately decodable subset is required to be decoded in order to completely decode the subset of the image; wherein each separately decodable subset is an image slice; each region map includes multiple entries, including one for each slice for the encoded field or frame, where each entry indicates whether the associated slice includes image data corresponding to the independently encoded region associated with the region map; and each region map provides an indication of those slices that must be decoded in order to completely decode the independently encoded region from a field or frame encoded representation of the associated image without decoding other image slices of the field or frame encoded representation. - View Dependent Claims (2, 3, 4)
-
-
5. For use with a predictive field or frame based encoded bitstream, each field or frame having multiple decodable subsets that can independently be recognized directly from the bitstream an improvement comprising:
-
a region identification signal stored on machine readable media; said region identification signal identifying all decodable subsets that represent an independently encoded object appearing as a subset of a field or frame; wherein said signal is adapted to permit extraction of data representing the independently encoded object by decoding the identified subsets directly from the bitstream; at least one region map, corresponding to the region identification, the region map providing an indication for each and every separately decodable subset of the encoded field or frame representation of the image whether that separately decodable subset is required to be decoded in order to completely decode the subset of the image; Wherein multiple objects are independently encoded in the an image sequence signal, across a plurality of frames; a region identification, configured as a region map, is provided for each of the plurality of frames and for each independently encoded object appearing in a frame; and said improvement further includes a global region presence indication for each frame, which indicates which of globally defined independently encoded regions, each corresponding to one of tie multiple objects, appears in the particular frame. - View Dependent Claims (6)
-
-
7. For use with a compressed image sequence where individual image frames are encoded in a manner adapted for decoding by a single decoder, an apparatus comprising:
-
machine readable media; and a signal stored on the machine readable media, said signal including a region map corresponding to an independently encoded region within a field or frame based encoding of an image, the independently encoded region representing a subset of both image data and compression data for the field or frame based encoding of the image; wherein the region map provides a mapping of the independently encoded region against all compression data for the field or frame based encoding of the image, said mapping adapted to permit the identification and selective extraction of the independently encoded regions without necessarily decompressing all compression data for the field or frame based encoding of the image; wherein multiple objects arc independently encoded in the image sequence across a plurality of frames; a region identification configured as a region map, is provided for each of the plurality of frames and for each independently encoded object appearing in a frame; and a global region presence indication for each frame, which indicates which of globally defined independently encoded regions, each corresponding to one of the multiple objects, appears in the particular frame. - View Dependent Claims (8, 9, 10)
-
-
11. An apparatus adapted for use in decompressing a frame-based or field-based dependent representation of an image within a compressed image sequence, the sequence including at least one anchor field or frame based representation that the dependent representation depends upon for predictive reconstruction, said apparatus comprising:
-
a signal stored on machine readable media; said signal including at least one region map for the anchor field or frame and at least one associated region identifier for the compressed dependent representation; wherein the region map for the anchor field or frame links a subset of image data represented by the anchor field or frame representation with a subset of compression data of the anchor representation; wherein the associated region map for the dependent representation links a subset of the image data represented by dependent representation with a subset of compression data of the dependent representation; wherein said signal is adapted to permit selective extraction and decoding of the subset of image data from the compressed dependent representation, without necessarily extracting and decoding all image data for the compressed dependent representation and the anchor field or frame, by extracting just the subset of compression data from the compressed dependent representation and the subset of compressed data from the anchor field and frame, using the region maps; and wherein each region map identifies all separately bitstream decodable subsets of image data utilized in the associated representation of compression data and identifies whether each separately bitstream decodable subset must be decoded in order to decode the linked subset of image data; wherein each separately decodable subset is an image slice of a field or frame; each region map includes multiple entries, including one for each slice of the encoded field or frame, where each entry indicates whether the associated slice includes image data corresponding to the independently encoded region corresponding to the region map; and each region map provides an indication of those slices that must be decoded in order to extract the independently encoded region from a field or frame based encoding of the associated image without decoding, other image slices of the field or frame based encoding. - View Dependent Claims (12, 13)
-
-
14. A method of enabling selective extraction and decompression of objects or regions of image data from within a field-based or frame-based encoding scheme for an image sequence, said method comprising:
-
identifying a subset of compression data representing a dependent field-based or frame-based encoding of a first image, where the subset has been independently encoded so as to depend for predictive reconstruction upon an associated subset of compression data representing a reference field-based or frame based encoding of a second image and so as to not depend for predictive reconstruction upon other image data found within the anchor field or frame; automatically creating a region map for each of the compression data for the first image and the compression data for the second image, where each region map identifies all compression data representing the field-based or frame-based encoding of the associated image, and demarks within that compression data boundaries of the subset, visible from the field-based or object-based encoding; wherein each region map is adapted to permit selective extraction and decompression of only the subsets, notwithstanding field-based or frame-based encoding of an entire associated image, without necessarily decoding each entire field or frame representation. - View Dependent Claims (15, 16)
-
-
17. An apparatus comprising:
-
machine readable media; a plurality of fields or frames in an image sequence stored on the machine readable media wherein multiple objects in an image in each field or frame are independently encoded; and a region map identifying all decodable subsets that represent each independently encoded object appearing as a subset of a field or frame for each field or frame, wherein a region map provides an indication for each and every separately decodable subset of the field or frame whether that separately decodable subset is required to be decoded in order to completely decode the subset of an image representing one object of the multiple objects and an indication of whether that subset is included in an independently encoded object for each of the other multiple objects. - View Dependent Claims (18)
-
Specification