Abstract: Classifying video for encoding optimization may include computing a content complexity score of a video, the content complexity score indicating a measure of how detailed the video is in terms of spatial and temporal information, categorizing the video into one of a plurality of buckets according to the content complexity score, each bucket representing a category of video content having a different range of content complexity scores and being associated with a ladder specific to the range, and encoding the video according to the ladder of the one of the plurality of buckets into which the video is categorized.

Abstract: Classifying video for encoding optimization may include computing a content complexity score of a video, the content complexity score indicating a measure of how detailed the video is in terms of spatial and temporal information, categorizing the video into one of a plurality of buckets according to the content complexity score, each bucket representing a category of video content having a different range of content complexity scores and being associated with a ladder specific to the range, and encoding the video according to the ladder of the one of the plurality of buckets into which the video is categorized.

Abstract: Macroblocking artifacts are assessed in an encoded video. Edges are detected and edge strengths are computed at macroblock boundaries in each decoded frame of the video. The detected edges are paired at horizontal or vertical sides of each macroblock to identify macroblock edge pairs. A macroblocking level of the frame is identified according to the macroblock edge pairs. For instance, for each of the macroblock edge pairs, the edges of the pair are identified as being macroblocking edges if edge strengths of both sides of the pair exceed a significance threshold, and the macroblocking level of the frame is determined by aggregating the macroblocking edges in the frame.

Abstract: Candidate encodes of a source video are produced using a plurality of encoding configurations. Quality of experience (QoE) scores for each of the candidate encodes are computed. The QoE scores of the plurality of candidate encodes are compared to determine a lowest bitrate encode that achieve a target QoE score. The lowest bitrate encode that achieve the target QoE score as an optimized output video is selected.

Abstract: Frames of a video stream collected at first and second points along a video delivery chain are buffered into first and second buffers, respectively, the second point being downstream the video delivery chain from the first point. A control group is identified as a subset of frames of the second buffer. Correlations of extracted features of the control group to extracted features of successive windows of frames of the first buffer are computed, the extracted features being based on spatial information and temporal information of the video stream. A delay between the video stream collected at the first point and the video stream collected at the second point is identified according to a maximum correlation of the correlations.

Abstract: Real-time latency of audio/video streams is identified. Signatures of a reference audio/video stream and signatures of a test audio/video stream are buffered. A needle is constructed as a vector including a set of signatures of the reference audio/video stream. Correlations of the needle to successive vectors of sets of signatures of the test audio/video stream are computed using a correlation function that calculates relatedness of the needle vector to each of the successive vectors of the test audio/video stream. A synchronization offset is identified between the test stream and the reference stream according to a maximum correlation point of the correlations of the needle to the successive sets of signatures of the test audio/video stream. The reference audio/video stream and the test audio/video stream are aligned according to the synchronization offset.

Abstract: A source quality of a source video and a source content complexity of the source video are identified. Parameter constraints with respect to parameters of an operation are received. The source video quality, source content complexity, and parameter constraints are applied to a deep neural network (DNN) producing DNN outputs. In an example, the DNN outputs are combined using domain knowledge to provide the filter parameters, as predicted, to a filter chain, such that applying the filter chain to the input source video results in an output video achieving the full reference video quality score. In another example, the DNN outputs are combined using domain knowledge to provide the filter parameters, as predicted, to a filter chain, such that applying the filter chain to the input source video results in an output video achieving the full reference video quality score.

Abstract: Correlated quality-of-experience (QoE) and latency measures are generated at a plurality of monitoring points along a multimedia delivery chain including multiple video operations. At each of the plurality of monitoring points, an absolute QoE measure defined on a human perceptual quality scale for media content is computed, and one or more of content extraction or feature extraction on the media content are performed. To a common middleware from each of the plurality of monitoring points, the respective QoE measure and results of the one or more of content extraction or feature extraction are transmitted. Computing and updating an absolute QoE measure for each of the plurality of monitoring points is performed. Computing and updating latencies between multiple monitoring points at the middleware using the results from each of the plurality of monitoring points is also performed.

Abstract: Banding effects in an image or video are assessed. The image or each frame of a video is decomposed into luminance or color channels. Signal activity is computed at each spatial location in each channel. A significance of the signal activity at each spatial location is determined by comparing each element of the signal activity with a significance threshold. Banding pixels are detected as those pixels of the image or frame having significant signal activity at their respective spatial locations and having non-significant signal activity of at least a minimum threshold percentage of neighboring pixels to the respective spatial locations.

Abstract: Correlated quality-of-experience (QoE) and latency measures are generated at a plurality of monitoring points along a multimedia delivery chain including multiple video operations. At each of the plurality of monitoring points, an absolute QoE measure defined on a human perceptual quality scale for media content is computed, and one or more of content extraction or feature extraction on the media content are performed. To a common middleware from each of the plurality of monitoring points, the respective QoE measure and results of the one or more of content extraction or feature extraction are transmitted. Computing and updating an absolute QoE measure for each of the plurality of monitoring points is performed. Computing and updating latencies between multiple monitoring points at the middleware using the results from each of the plurality of monitoring points is also performed.

Abstract: The present invention is a system or method that facilitates smart decision makings at the client side for adaptive video streaming over a video delivery network by making use of perceptual video quality-of-experience predictions performed during the video preparation stage, at the video hosting or cache server side, or inside the video delivery network, and then transmitted to the client. Compared with prior art approaches of adaptive bitrate video streaming, the present invention can result in one or more of the following benefits: 1) save the overall bandwidth for the delivery of the video content without scarifying the client users' quality-of-experience; 2) create better overall visual quality-of-experience of the client users; 3) create smoother visual quality-of-experience of the client users; and 4) reduce the probability of rebuffering or stalling events at the client user side.

Abstract: Disclosed are system and method that auto-mate measurement of end users' quality-of-experience (QoE) when perceiving the video being streamed to the users' viewing devices. The overall user QoE is measured and computed by combining the instantaneous presentation quality, the playback smoothness quality, and the interactions between them. Prediction accuracy is thus significantly improved. The instantaneous and end-of-process QoE measures created by the system and method described are suitable for the monitoring and optimization of media streaming systems and services.

Abstract: There is disclosed a method and system for objective video quality assessment. The objective video quality assessment approach automatically predicts human visual quality assessment behaviors in evaluating videos that contain artifacts generated during video acquisition, storage, reproduction, compression, transmission, processing, and/or display. The present method and system computes a five-dimensional quality map of a video being assessed, where the map indicates the local quality variations of the video in a five-dimensional space, which includes two spatial dimensions, one scale dimension, one time dimension, and one distortion type dimension. The present method and system may use one or a combination of device and viewing condition parameters in the generation of the quality map, and in the combination of the quality map into a scalar or vector-values measure that indicates quality aspects of the test videos.

Abstract: The present invention is a system and method for video coding. The video coding system may involve a structural similarity-based divisive normalization approach, wherein the frame prediction residual of the current frame may be transformed to form a set of coefficients and a divisive normalization mechanism may be utilized to normalize each coefficient. The normalization factor may be designed to reflect or approximate the normalization factor in a structural similarity definition. The Lagrange parameter for RDO for divisive normalization coefficients may be determined by both the quantization step and a prior distribution function of the coefficients. The present invention may generally be utilized to improve the perceptual quality of decoded video without increasing data rate, or to reduce the data rate of compressed video stream without sacrificing the perceived quality of decoded video. The present invention has shown to significantly improve the coding efficiency of MPEG4/H.