Method and device for compressing a video sequence

摘要

The present invention relates to the field of video compression and, specifically, to controlling the bit rate of the generated compressed sequence. This invention relates to a method for constant bit rate compression, using a step of normalizing the compression ratio. Variations in said ratio can thus be smoothed out, eliminating sudden variations in image quality in the resulting sequences. The perceived quality of the service is thereby improved.

主权项

1. Method of compressing a video sequence of images using an information processing apparatus in the form of a computer or a dedicated electronic compression module component, the method comprising:
using a computer or dedicated compression module component to first compress the video sequence at a constant bit rate and determine a compressed image size value and an associated compression factor value for each image in the sequence; using said computer or dedicated compression module component to compute normalized values of compression factor values associated with each compressed image such that a normalized compression factor value is optimized in accordance with a size of a particular receiver buffer to be used for decompressing the sequence, wherein computing normalized compression factor values comprises using said computer or dedicated compression module component to perform calculations for simulating a data filling level of the receiver buffer
(a) setting an initial filling level of a simulated receiver buffer to a predetermined initial value;(b) for each successive image in the video sequence, calculating a filling level based upon a size of the image, the associated compression factor and a predetermined rate of data consumption from said particular receiver buffer which would occur during decompression of the data;(c) in a case where a calculated filling level for a particular image in the video sequence causes neither overflow nor famine of the simulated receiver buffer, the associated compression factor is retained and a filling level for a next successive image in the video sequence is calculated;(d) in a case where a calculated filling level causes an overflow or famine within the simulated receiver buffer, the computer processor or compression module component further performs operations of:
(i) identifying an oldest image having an associated compression factor value which is a maximum or minimum amongst compression factor values of the images in the video sequence compressed at the constant bit rate;(ii) modifying the associated compression factor value of the identified oldest image so as to increase a quantity of image data to be decompressed in a case of famine within the buffer and decrease a quantity of image data to be decompressed in a case of overflow of the buffer; and(iii) beginning with said identified oldest image, perform a rain said calculations for simulating a data filling level of said particular receiver buffer for each successive image in the video sequence; and, using said computer or dedicated compression module component to compress the video sequence a second time using the computed normalized values of said compression factor value associated with each image.