Method and apparatus for adjusting stereoscopic image parallax
The present invention provides a method for adjusting stereoscopic image parallax. An original camera is provided. A scene space corresponding to the original camera is preset, including presetting a depth range of the scene space. Preset viewing environment parameters for displaying stereoscopic images of a scene in the scene space are configured, including presetting a depth range of an actual view space. The method further includes establishing a mapping relationship between the depth range of the actual view space and the depth range of the scene space. According to the mapping relationship, the preset viewing environment parameters and the scene space, camera parameters for adjusting stereoscopic image parallax are calculated. The original camera is adjusted based on the camera parameters to capture the scene. Thus, the stereoscopic images of the scene can be generated and presented.
1. A method for adjusting stereoscopic image parallax, comprising:
providing an original camera;
presetting a depth range of a scene space, wherein the scene space is corresponding to the original camera;
configuring preset viewing environment parameters for displaying stereoscopic images of a scene in the scene space, including presetting a depth range of an actual view space;
establishing a mapping relationship between the depth range of the actual view space and the depth range of the scene space;
according to the mapping relationship, the preset viewing environment parameters and the scene space, calculating camera parameters for adjusting the stereoscopic image parallax;
based on the camera parameters, adjusting the original camera to capture the scene; and
generating the stereoscopic images of the scene,
wherein:
the depth range of the scene space is preset by a depth of positive-parallax plane z1 and a depth of negative-parallax plane z2;
a left camera and a right camera are obtained by respectively moving the original camera to two sides in horizontal direction; and
calculating the camera parameters further comprises:
a first calculation step including: based on the preset viewing environment parameters, the preset depth of the positive-parallax plane z1, the preset depth of the negative-parallax plane z2 and current camera parameters, calculating a separation value between the left camera and the right camera sep, a depth of zero-parallax plane con, and a separation-to-parallax ratio ratio between the separation value sep and the depth of the zero-parallax plane con;
a second calculation step including: keeping one of ratio and sep unchanged, based on a depth of a target object closest to the original camera in a current scene, calculating a new separation value between the left camera and the right camera and a new depth of the zero-parallax plane, such that the current scene is captured using the new separation value and the new depth of the zero-parallax plane.
2. The method according to claim 1 , wherein:
the preset viewing environment parameters includes a distance D between a viewer and the stereoscopic images corresponding to a stereo display window, an interocular distance e of the viewer, a width of the stereo display window Tw, and the depth range of the actual view space; and
the depth range of the actual view space is preset by a maximum protruding distance CS and a maximum recess distance BS of the stereoscopic images.
3. The method according to claim 2 , wherein the first calculation step further comprises:
using a first preset algorithm to calculate a positive parallax value on the stereo display window w1 and a negative parallax value on the stereo display window w2 based on the preset viewing environment parameters, wherein: w1 is positively correlated with at least one of CS and e, w1 is negatively correlated with D−CS, w2 is positively correlated with at least one of BS and e, and w2 is negatively correlated with D+BS; and
using a second preset algorithm to calculate a positive parallax value on the view plane Vw1 and a negative parallax value on the view plane Vw2 based on w1 and w2, wherein: Vw1 is positively correlated with w1, Vw1 is negatively correlated with Tw, Vw2 is positively correlated with w2, and Vw2 is negatively correlated with Tw.
4. The method according to claim 3 , wherein the first calculation step further comprises:
provided that U denotes the width of a near projection plane of the original camera in the observation space, near denotes a distance between the near projection plane of the original camera and the coordinate origin in the observation space, z denotes depth of a point in the camera space, and Vw denotes a parallax value of the point on the view plane,
using a third preset algorithm to calculate the separation-to-parallax ratio ratio according to current camera parameters, z1 and z2, wherein the third preset algorithm uses the equation
ratio
=
U
near
*
(
2
*
(
z
2
-
z
1
)
Vw
1
*
z
1
-
V
w
2
*
z
2
-
1
)
;
using a fourth preset algorithm to calculate the depth of the zero-parallax plane con, wherein the fourth preset algorithm uses the equation
con
=
0.5
*
Vw
*
U
*
z
near
*
ratio
+
0.5
*
Vw
*
z
+
z
;
and
using a fifth preset algorithm to calculate the separation value between the left camera and the right camera sep based on ratio and con, wherein the fifth preset algorithm uses the equation
ratio
=
sep
con
.
5. The method according to claim 1 , wherein the second calculation step further comprises:
filtering a plurality of objects in a current scene according to object classifications;
sorting the objects within a projection volume of the original camera according to distances between the objects and the original camera;
excluding objects between a near projection plane of the original camera and a preset true near plane;
determining the target object closest to the original camera; and
calculating the depth of the target object.
6. The method according to claim 5 , wherein:
provided that U denotes the width of the near projection plane of the original camera in the observation space, near denotes a distance between the near projection plane of the original camera and the coordinate origin in the observation space, z denotes depth of a point in the camera space, and Vw denotes a parallax value of the point on the view plane,
the scene is a virtual scene, the original camera is a virtual camera used to render the virtual scene for generating the stereoscopic images of the virtual scene, and
the second calculation step further includes:
comparing the depth of the target object and the preset depth of the positive-parallax plane z1;
when the depth of the target object is less than z1, keeping ratio unchanged, assigning the depth of the target object as the preset depth of the positive-parallax plane, and according to a fourth preset algorithm and a fifth preset algorithm, calculating the new separation value between the left and right cameras and the new depth of the zero-parallax plane; and
when the depth of the target object is greater than z1, keeping ratio unchanged, based on z1, the fourth preset algorithm and the fifth preset algorithm, calculating the new separation value between the left and right cameras and the new depth of the zero-parallax plane;
wherein the fourth preset algorithm uses the equation
con
=
0.5
*
Vw
*
U
*
z
near
*
ratio
+
0.5
*
Vw
*
z
+
z
and the fifth preset algorithm uses the equation
ratio
=
sep
con
.
7. The method according to claim 5 , wherein:
the scene is a virtual scene, the original camera is a virtual camera used to render the virtual scene for generating the stereoscopic images of the virtual scene, and
the second calculation step further includes:
comparing the depth of the target object and the preset depth of the positive-parallax plane z1;
when the depth of the target object is less than z1, keeping sep unchanged, using the depth of the target object as the preset depth of the positive-parallax plane, and according to a sixth preset algorithm, calculating the new depth of the zero-parallax plane to render the virtual scene; and
when the depth of the target object is greater z1, directly using the separation value sep and the depth of the zero-parallax plane con to render the virtual scene.
8. The method according to claim 5 , wherein:
the scene is a real scene, the original camera is a stereo camera for capturing stereoscopic images of the real scene; and
the second calculation step further includes:
using a seventh preset algorithm to calculate a rotating angle between the left and right cameras and z-axis of the original camera based on the separation-to-parallax ratio, wherein the rotating angle is positively correlated with the separation-to-parallax ratio;
comparing the depth of the target object with the preset depth of the positive-parallax plane z1; and
when the depth of the target object is less than z1,
keeping sep unchanged, using the depth of the target object as the preset depth of the positive-parallax plane,
according to a fifth preset algorithm and a sixth preset algorithm, calculating a new separation-to-parallax ratio and the new depth of the zero-parallax plane;
calculating a new rotating angle according to the seventh preset algorithm; and
adjusting the left and right cameras according to the new rotating angle.
9. The method according to claim 1 , before the second calculation step, further comprising:
comparing the separation-to-parallax ratio with a preset ratio threshold;
when the separation-to-parallax ratio is less than the preset ratio threshold, permitting to perform the second calculation step; and
when the separation-to-parallax ratio is greater than the preset ratio threshold, generating a reminder for a user to reconfigure at least one of the preset viewing environment parameters, the preset depth of the positive-parallax plane and the preset depth of the negative-parallax plane.
10. An apparatus for adjusting stereoscopic image parallax, comprising:
a mapping relationship establishing unit configured to establish a mapping relationship between a depth range of an actual view space and a depth range of a scene space,
wherein:
an original camera is provided to capture a scene in the scene space,
according to the mapping relationship, camera parameters are calculated for adjusting the stereoscopic image parallax,
based on the camera parameters, the original camera is adjusted to capture the scene such that stereoscopic images of the scene are correspondingly generated, and
the depth range of the scene space is preset by a depth of positive-parallax plane z1 and a depth of negative-parallax plane z2, and any portions of objects located between the positive-parallax plane z1 and the negative-parallax plane z2 in the scene space are rendered to satisfy requirements for maximum protruding and maximum recess display effects; and
a first calculation unit configured to: based on preset viewing environment parameters, the preset depth of the positive-parallax plane z1, the preset depth of the negative-parallax plane z2 and current camera parameters, calculate a separation value between the left camera and the right camera sep, a depth of zero-parallax plane con, and a separation-to-parallax ratio ratio between the separation value sep and the depth of the zero-parallax plane con; and
a second calculation unit configured to: keeping one of ratio and sep unchanged, based on a depth of a target object closest to the original camera in a current scene, calculate a new separation value between the left camera and the right camera and a new depth of the zero-parallax plane, such that the current scene is captured using the new separation value and the new depth of the zero-parallax plane;
wherein: the depth range of the scene space is preset by z1 and z2; and the left camera and the right camera is obtained by respectively moving the original camera to two sides in horizontal direction.
11. The apparatus according to claim 10 , wherein:
the preset viewing environment parameters includes a distance D between a viewer and the stereoscopic images corresponding to a stereo display window, an interocular distance e of the viewer, a width of the stereo display window Tw, and the depth range of the actual view space; and
the depth range of the actual view space is preset by a maximum protruding distance CS and a maximum recess distance BS of the stereoscopic images.
12. The apparatus according to claim 11 , wherein the first calculation unit further comprises:
a first parallax calculation unit configured to apply a first preset algorithm to calculate a positive parallax value on the stereo display window w1 and a negative parallax value on the stereo display window w2 based on the preset viewing environment parameters, wherein: w1 is positively correlated with at least one of CS and e, w1 is negatively correlated with D−CS, w2 is positively correlated with at least one of BS and e, and w2 is negatively correlated with D+BS; and
a second parallax calculation unit configured to apply a second preset algorithm to calculate a positive parallax value on the view plane Vw1 and a negative parallax value on the view plane Vw2 based on w1 and w2, wherein: Vw1 is positively correlated with w1, Vw1 is negatively correlated with Tw, Vw2 is positively correlated with w2, and Vw2 is negatively correlated with Tw.
13. The apparatus according to claim 12 , wherein the first calculation unit further comprises:
provided that U denotes the width of a near projection plane of the original camera in the observation space, near denotes a distance between the near projection plane of the original camera and the coordinate origin in the observation space, z denotes depth of a point in the camera space, and Vw denotes a parallax value of the point on the view plane,
a ratio calculation unit configured to apply a third preset algorithm to calculate the separation-to-parallax ratio ratio according to current camera parameters, z1 and z2, wherein the third preset algorithm uses the equation
ratio
=
U
near
*
(
2
*
(
z
2
-
z
1
)
Vw
1
*
z
1
-
Vw
2
*
z
2
-
1
)
;
a depth calculation unit configured to apply a fourth preset algorithm to calculate the depth of the zero-parallax plane con, wherein the fourth preset algorithm uses the equation
con
=
0.5
*
Vw
*
U
*
z
near
*
ratio
+
0.5
*
Vw
*
z
+
z
;
and
a separation calculation unit configured to apply a fifth preset algorithm to calculate the separation value between the left camera and the right camera sep based on ratio and con, wherein the fifth preset algorithm uses the equation
ratio
=
sep
con
.
14. The apparatus according to claim 10 , wherein the second calculation unit further comprises:
a classification unit configured to filter a plurality of objects in a current scene according to object classifications;
an object exclusion unit configured to sort the objects within a projection volume of the original camera according to distances between the objects and the original camera, and to exclude objects between a near projection plane of the original camera and a preset true near plane; and
an object depth calculation unit configured to determine the target object closest to the original camera and to calculate the depth of the target object.
15. The apparatus according to claim 14 , wherein:
provided that U denotes the width of the near projection plane of the original camera in the observation space, near denotes a distance between the near projection plane of the original camera and the coordinate origin in the observation space, z denotes depth of a point in the camera space, and Vw denotes a parallax value of the point on the view plane,
the scene is a virtual scene, the original camera is a virtual camera used to render the virtual scene for generating the stereoscopic images of the virtual scene, and the second calculation unit further comprises:
a comparison unit configured to compare the depth of the target object and the preset depth of the positive-parallax plane z1; and
a secondary calculation unit configured to:
when the depth of the target object is less than z1, keep ratio unchanged, assign the depth of the target object as the preset depth of the positive-parallax plane, and according to a fourth preset algorithm and a fifth preset algorithm, calculate the new separation value between the left and right cameras and the new depth of the zero-parallax plane; and
when the depth of the target object is greater than z1, keeping ratio unchanged, and based on z1, the fourth preset algorithm and the fifth preset algorithm, calculate the new separation value between the left and right cameras and the new depth of the zero-parallax plane;
wherein the fourth preset algorithm uses the equation
con
=
0.5
*
Vw
*
U
*
z
near
*
ratio
+
0.5
*
Vw
*
z
+
z
;
and the fifth preset algorithm uses the equation
ratio
=
sep
con
.
16. The apparatus according to claim 14 , wherein:
the scene is a virtual scene, the original camera is a virtual camera used to render the virtual scene for generating the stereoscopic images of the virtual scene, and
the second calculation unit further comprises
a comparison unit configured to compare the depth of the target object and the preset depth of the positive-parallax plane z1; and
a secondary calculation unit configured to
when the depth of the target object is less than z1, keep sep unchanged, use the depth of the target object as the preset depth of the positive-parallax plane, and according to a sixth preset algorithm, calculate the new depth of the zero-parallax plane to render the virtual scene; and
when the depth of the target object is greater than z1, directly use the separation value sep and the depth of the zero-parallax plane con to render the virtual scene.
17. The apparatus according to claim 5 , wherein:
the scene is a real scene, the original camera is a stereo camera for capturing stereoscopic images of the real scene; and
the second calculation unit further comprises:
an angle calculation unit configured to use a seventh preset algorithm to calculate a rotating angle between the left and right cameras and z-axis of the original camera based on the separation-to-parallax ratio, and to compare the depth of the target object and the preset depth of the positive-parallax plane z1, wherein the rotating angle is positively correlated with the separation-to-parallax ratio; and
a secondary calculation unit configured to: when the depth of the target object is less z1,
keep sep unchanged, use the depth of the target object as the preset depth of the positive-parallax plane;
calculate a new separation-to-parallax ratio and the new depth of the zero-parallax plane according to the fifth preset algorithm and the sixth preset algorithm; and
calculate a new rotating angle according to the seventh preset algorithm, wherein the left and right cameras are adjusted according to the new rotating angle.
18. The apparatus according to claim 10 , further comprising a reminder unit configured to:
compare the separation-to-parallax ratio and a preset ratio threshold,
permit the second calculation unit to operate when the separation-to-parallax ratio is less than the preset ratio threshold; and
when the separation-to-parallax ratio is greater than the preset ratio threshold, generate a reminder for a user to reconfigure at least one of the preset viewing environment parameters, the preset depth of the positive-parallax plane and the preset depth of the negative-parallax plane.