Microsoft Technology Licensing, LLC

Walid S I Ali

Rony Ferzli

Ming-Chieh Lee

Video encoding computations are optimized by dynamically adjusting slice patterns of video frames based on complexity of each frame and allocating multi-core threading based on the slices. The complexity may be based on predefined parameters such as color, motion, and comparable ones for each slice. Allocation is determined based on capacity and queue of each processing core such that overall computation performance for video encoding is improved.

Optimized allocation of multi-core computation for video encoding

A block-based, no-reference sharpness metric is provided taking advantage of Human Visual System (HVS) characteristics. Texture and smooth region blocks are excluded in computing the metric since sharpness is perceived mostly around edges. Overall sharpness metric is computed by pooling simulated combination of information in human brain employing a logistic function to replicate the behavior of HVS.

Mapping psycho-visual characteristics in measuring sharpness feature and blurring artifacts in video streams

A low complexity block-based, no-reference objective blockiness metric is provided that may be combined with other artifact metrics to measure overall quality of received video stream in a video conferencing application such that measures can be taken at the transmitter or in post-processing to enhance video quality. Prior knowledge of the blockiness boundaries may be used to reduce number of computations in determining the blockiness of a particular video frame.

Growing community of inventors

Sammamish, WA, United States of America

Walid S I Ali