Content-dependent video quality model for video streaming services
AssigneeDeutsche Telekom AG
InventorsMarie-Neige Garcia, Alexander Raake, Savvas Argyropoulos, Bernhard Feiten, Peter List
Life of this patent
Estimating the term…
Abstract
A method for estimating the perception quality of a digital video signal includes: (1a) extracting information of the video bit stream, which is captured prior to decoding; (1b) getting estimation(s) for one or more impairment factors IF using, for each of the estimations, an impact function adapted for the respective impairment factor; and (1c) estimating the perceived quality of the digital video signal using the estimation(s) obtained in step (1b).
Claims31
A method for estimating the perception quality of a digital video signal, the method comprising the steps of:
- (1a) extracting information of the video bit stream, which is captured prior to decoding;
- (1b) getting estimation(s) for one or more impairment factors IF using, for each of the estimations, an impact function adapted for the respective impairment factor;
- (1c) estimating the perceived quality of the digital video signal using the estimation(s) obtained in step (1b);
- wherein each of the impact functions used in step (1b) takes as input a set of content-dependent parameters q computed from a set of Group Of Picture (GOP)/scene-complexity parameters, wherein the GOP/scene-complexity parameters are derivable from packet-header information and available in case of encrypted video bit streams;
- wherein the set of content-dependent parameters q is derived at least from a GOP/scene-complexity parameter SscI, denoting the average I frame size per scene;
- and
- wherein for estimating at least one of the impairment factors, an impact function ƒIF is used that depends on a content-dependent parameter q1 being computed from the reciprocal of the weighted mean of the GOP/scene-complexity parameter SscI over the scenes sc multiplied by a coefficient;
- and
- wherein each scene sc has a weight of wsc×Nsc with Nsc being the number of GOPs per scene and wsc being a weight factor, wherein for the scenes having the lowest SscI value:
- wsc is set to a value greater than 1, and for all other scenes:
- wsc is set equal to 1.
An apparatus for estimating the perception quality of a digital video signal, the apparatus comprising:
- a processor, configured for extracting information from a video bit stream being captured prior to decoding;
- at least one impact estimator;
- a quality estimator configured for estimating the perception quality Qv of the video signal;
- wherein each of the impact estimator(s) is configured for estimating the quality impact due to an impairment factor by an impairment function taking as input a set of content-dependent parameters q computed from a set of Group Of Picture (GOP)/scene-complexity parameters, wherein the GOP/scene-complexity parameters are derivable from packet-header information and available in case of encrypted video bit streams;
- wherein the set of content-dependent parameters q is derived at least from a GOP/scene-complexity parameter SscI, denoting the average I frame size per scene;
- wherein for estimating at least one of the impairment factors an impact function ƒIF is used that depends on a content-dependent parameter q1 being computed from the reciprocal of the weighted mean of the GOP/scene-complexity parameter SscI over the scenes sc multiplied by a coefficient;
- and
- wherein each scene sc has a weight of wsc×Nsc with Nsc being the number of GOPs per scene and wsc being a weight factor, wherein or the scenes having the lowest SscI value:
- wsc is set to a value greater than 1, and for all other scenes:
- wsc is set equal to 1.
Dependencies are inferred from the claim text.