Full reference system for predicting subjective quality of three-dimensional video

摘要

A method of generating a predictive picture quality rating makes a disparity measurement of a three-dimensional image by comparing left and right sub-components of the three-dimensional image. Then the left and right sub-components of the three-dimensional image are combined (fused) into a two-dimensional image, using data from the disparity measurement for the combination. A predictive quality measurement is then generated based on the two-dimensional image, and further including quality information about the comparison of the original three-dimensional image.

主权项

1. A method of generating a predictive picture quality rating, comprising:
accepting a three-dimensional image having a left portion and a right portion; generating a disparity measurement by comparing the left and right portions of the three-dimensional image; combining the left and right portions of the three-dimensional image into a single two-dimensional image using the disparity measurement; accepting a second three-dimensional image having a left portion and a right portion; generating a second disparity measurement by comparing the left and right portions of the second three-dimensional images; combining the left and right portions of the second three-dimensional image into a single second two-dimensional image using the second disparity measurement; applying at least a first predictive quality technique to the two-dimensional image and to the second two-dimensional image to generate a two-dimensional quality rating; modifying the two-dimensional quality rating with quality information about the three-dimensional image or the second three-dimensional image to generate an overall quality rating; and outputting the overall quality rating.