Video processing method and apparatus for use with a sequence of stereoscopic images
View Patent ↗To generate a warning that a stereoscopic image sequence has been synthesized from a 2D image sequence, a video processor correlates left-eye image data and right-eye image data to identify any sustained temporal offset between the left-eye and right-eye image data. A measure of sustained correlation between a measured spatial distribution of horizontal disparity and a spatial model can also be used to generate the warning.
1. A video processing method comprising the steps in a video processor of:
receiving left-eye image data and right-eye image data from a sequence of stereoscopic images;
performing a correlation process upon said left-eye image data and right-eye image data to identify any sustained temporal offset between the said left-eye and right-eye image data; and
performing an analysis of said sustained temporal offset to generate a warning that the said stereoscopic image sequence has been derived from a 2D image sequence,
where a spatial distribution of horizontal disparity between said left-eye and right-eye image data is measured and where a measure of sustained correlation between the measured spatial distribution of horizontal disparity between the said left-eye and right-eye image data and a spatial model of the said spatial distribution is used to generate a warning that the said stereoscopic image sequence has been derived from a 2D image sequence.
2. A method according to claim 1 in which the said measure of correlation is evaluated for left-eye and right-eye data which has been time-aligned using said identified sustained temporal offset.
3. A method according to claim 1 in which a combination of sustained temporal offset and sustained correlation with a spatial disparity model is used to generate the said warning.
4. A method according to claim 1 in which the said spatial disparity model is a linear model derived from linear regression of measured disparity values between time-aligned left-eye and right-eye data.
5. A method of analysing a sequence of stereoscopic images comprising left-eye and right-eye image data, comprising the steps in a video processor of:
forming a measure of spatial distribution of horizontal disparity between the said left-eye and right-eye image data;
forming a measure of sustained correlation between said measured spatial distribution of horizontal disparity and a spatial model of the said spatial distribution; and
using said measure of sustained correlation to generate a warning that the said stereoscopic image sequence has been derived from a 2D image sequence; in which:
said measure of sustained correlation is evaluated for left-eye and right-eye data which has been time-aligned by a correlation process operating on mutually delayed left and right eye data to measure a sustained temporal offset; and
a combination said measure of sustained temporal offset and said measure of sustained correlation with a spatial disparity model is used to generate the said warning.
6. A method according to claim 5 in which the said spatial disparity model is a linear model derived from linear regression of measured disparity values between time-aligned left-eye and right-eye data.
7. Video processing apparatus comprising:
an input for receiving left-eye image data and right-eye image data from a sequence of stereoscopic images;
a correlator for performing a correlation process upon said left-eye image data and right-eye image data to identify any sustained temporal offset between the said left-eye and right-eye image data; and
a logic block for performing an analysis to generate a warning that the said stereoscopic image sequence has been derived from a 2D image sequence;
where said logic block utilises a measure of sustained temporal offset between the said left-eye and right-eye image data to generate a warning that the said stereoscopic image sequence has been derived from a 2D image sequence; and
where said logic block serves to form a measure of the spatial distribution of horizontal disparity between the said left-eye and right-eye image data and to identify any sustained correlation between said measured spatial distribution of horizontal disparity between the said left-eye and right-eye image data and a spatial model of the said spatial distribution to generate said warning.
8. Apparatus according to claim 7 in which the said measure of correlation is evaluated for time-aligned left-eye and right-eye data.
9. Apparatus according to claim 7 in which a combination of sustained temporal offset and sustained correlation with a spatial disparity model is used to generate the said warning.
10. Apparatus according to claim 7 in which the said spatial disparity model is a linear model derived from linear regression of measured disparity values between time-aligned left-eye and right-eye data.