Analyzer and methods for detecting and processing video data types in a video data stream
First Claim
1. A method for improving encoding efficiency of a video data stream, said method comprising:
- analyzing each field in said video data stream to detect repeated fields wherein said video data stream includes combinations of a video sequence, a cartoon sequence, and a telecine sequence and further wherein repeated fields in said cartoon sequence are detected and repeated fields in said telecine sequence are detected; and
dropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency.
11 Assignments
0 Petitions
Accused Products
Abstract
A video data stream analyzer modifies an input digital video signal so that the resulting output digital signal can be optimally compressed by a digital video encoder. The video data stream analyzer eliminates redundancy in the input video signal, and reorganizes the input video signal so that the spatial and temporal redundancy is increased. In addition, the video data stream analyzer generates side channel information that is supplied to the video encoder. The side channel information tells the video encoder whether vertical frame-based filtering or vertical field-based filtering is preferable. Additional side channel information specifies the order and duration of the display of the fields after decoding and this information preferably is encoded with the video signal. The video data stream analyzer provides scan detection of the incoming video digital data, and automatically and reliably detects scene cuts, repeated fields, and mixed-field frames in the incoming digital video data in real time independent of the video source. The video data stream analyzer modifies the input video data stream by dropping repeated fields and replacing a frame with a scene cut with a frame having identical fields for video, cartoon, telecine video sources as well as arbitrary combinations of these video sources.
120 Citations
51 Claims
-
1. A method for improving encoding efficiency of a video data stream, said method comprising:
-
analyzing each field in said video data stream to detect repeated fields wherein said video data stream includes combinations of a video sequence, a cartoon sequence, and a telecine sequence and further wherein repeated fields in said cartoon sequence are detected and repeated fields in said telecine sequence are detected; and dropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35)
-
-
36. A method for improving encoding efficiency of a video data stream including a plurality of frames and each frame includes a field of a first parity and a field of a second parity, said method comprising:
-
analyzing each field in said video data stream to detect repeated fields wherein said analyzing process further comprises; comparing a field activity of a field with a first threshold; and identifying said field as a repeated field if said field activity of said field is less than said first threshold and said field has said second parity; comparing a strip activity of each strip of said field with a second threshold; and identifying said field as a repeated field if a predetermined number of said strip activities of said field are less than said second threshold and sail field has said second parity; repeating said field activity and strip activity comparisons for a second field of said second parity if (i) said field has said first parity;
(ii) said field activity of said field is less than said first threshold; and
(iii) a predetermined number of said strip activities of said field are less than said second threshold;identifying said field as a repeated field if (i) said field activity of said second field is greater than said first threshold;
or (ii) a predetermined number of said strip activities of said second field are greater than said second threshold; anddropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency. - View Dependent Claims (40, 41, 42, 43)
-
-
37. A method for improving encoding efficiency of a video data stream including a plurality of frames and each frame includes a field of a first parity and a field of a second parity, said method comprising:
-
analyzing each field in said video data stream to detect repeated fields wherein said analyzing process further comprises; comparing a field activity of a field with a first threshold; and identifying said field as a repeated field if said field activity of said field is less than said first threshold and said field has said second parity; comparing a strip activity of each strip of said field with a second threshold; and identifying said field as a repeated field if a predetermined number of said strip activities of said field are less than said second threshold and said field has said second parity; repeating said field activity and strip activity comparisons for a second field of said second parity if (i) said field has said first parity;
(ii) said field activity of said field is less than said first threshold; and
(iii) a predetermined number of said strip activities of said field are less than said threshold;comparing a measure of field activity of said field with a third threshold if (i) said field activity of said second field is less than said first threshold; and
(ii) a predetermined number of said strip activities of said second field are less than said second threshold; andidentifying said field as a repeated field if said measure of field activity of said field is less than said third threshold; and dropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency.
-
-
38. A method for improving encoding efficiency of a video data stream including a plurality of frames and each frame includes a field of a first parity and a field of a second parity, said method comprising:
-
analyzing each field in said video data stream to detect repeated fields wherein repeated fields in a telecine sequence in said video data stream are detected and repeated fields in a cartoon sequence in said video data stream are detected and further wherein said analyzing process further comprises; comparing a field activity of a field with a first threshold; and identifying said field as a repeated field if said field activity of said field is less than said first threshold and said field has said second parity; and dropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency. - View Dependent Claims (39)
-
-
44. A method for improving encoding efficiency of a video data stream, said method comprising:
-
analyzing each field in said video data stream to detect repeated fields wherein said analyzing each field further comprises; measuring a strip activity between a strip in a first field of a first parity and a strip of a second field of said first parity in said video data steam for a plurality of strips thereby obtaining a plurality of strip activities for said first field; generating a field activity for said first field using said plurality of strip activities; comparing said field activity to a first threshold; comparing each of said plurality of strip activities to a second threshold; and identifying said first field as a repeated field if (i) said first parity is an even parity;
(ii) said field activity for said first field is less than said first threshold; and
(iii) a predetermined number of said strip activities of said first field are less than said second threshold; anddropping each detected repeated field from said video data stream to generate a modified video data stream thereby eliminating encoding of redundant information and so improving encoding efficiency. - View Dependent Claims (45, 46, 47, 48, 49, 50, 51)
-
Specification