Method and device for tone-mapping a picture by using a parametric tone-adjustment function
The present principles relates to a method and device for tone-mapping an input picture by using a parametric tone-adjustment function. The method is characterized in that the method comprises determining at least one parameter of said tone-adjustment function modulated by a brightness level of the input picture.
1. A method for tone-mapping an input picture by using a parametric tone-adjustment function described over multiple ranges, at least one parameter of the tone-adjustment function being associated with each range, wherein the method comprises:
determining a non-linear tone adjustment function over at least one range according to a modulation value responsive to the brightness level; and
determining a parameter of the tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the input picture.
2. The method of claim 1 , wherein the method further comprises determining another parameter of the tone-adjustment function according to an increasing concave function taking into account a brightness level of the input picture.
3. The method of claim 1 , wherein the tone adjustment function is linear over a first range, parabolic over a second range and linear over a third range.
4. The method of claim 3 , wherein the tone adjustment function is described:
over the first range by:
y=SGC*x
where SGC is a real parameter value, y is an output value, and x is an input value;
over the second range by:
ax 2 +bx+c,x SGC <x<x HGC
over the third range by:
y=HGC*x +(1− HGC )
where HGC is a slope; and
where
{
a
=
-
0.5
·
SGC
-
HGC
MTA
b
=
1
-
HGC
MTA
+
SGC
+
HGC
2
c
=
1
-
HGC
SGC
-
HGC
+
MTA
2
{
x
SGC
=
1
-
HGC
SGC
-
HGC
-
MTA
2
x
HGC
=
1
-
HGC
SGC
-
HGC
+
MTA
2
where MTA is a real parameter value and
{
exposure
=
shadowGainControl
4
+
0
,
5
expgain
=
v
(
L
source
L
target
)
with
v
a
function
based
on
L
target
SGC
=
expgain
.
exposure
HGC
=
highLightGainControl
4
MTA
=
midTonesAdjustment
2
with
v
(
x
)
=
log
(
1
+
(
ρ
-
1
)
.
x
1
γ
)
log
(
ρ
)
and
ρ
=
1
+
32.
(
L
target
100000
)
1
/
γ
where shadowGainControl and midTonesAdjustement are real values given by:
{
shadowGainControl
=
0.229
+
2.47
*
exp
(
-
0.333
*
Ba
)
midTonesAdjustment
=
2.21
*
exp
(
-
0.16
*
Ba
)
where highlightGainControl is a real value, and a modulation value Ba is responsive to a brightness level of the input picture.
5. The method of claim 4 , wherein the real value highlightGainControl is a constant value.
6. The method of claim 4 , wherein the real value highlightGainControl is kept as another parameter in addition to the modulation value.
7. The method of claim 4 , wherein the real value highlightGainControl is given by:
highlightGainControl=2−2*exp(−0.26/ Ba ).
8. The method of claim 4 , wherein the modulation value is responsive to a mid-tone level of the input picture.
9. The method of claim 8 , wherein the mid-tone level is determined based on a black level and a white level.
10. The method of claim 9 , wherein the mid-tone level is determined as one of a geometric mean and a logarithm mean of the black level and the white level.
11. The method of claim 4 , wherein the input image is one of a plurality of pictures included in a video sequence, and wherein the determining of the modulation value is performed for each of the plurality of pictures, wherein the modulation values for the plurality of pictures are temporally smoothed.
12. The method of claim 3 , wherein the tone adjustment function is described:
over the first range by:
y
=
SGC
′
*
x
Ba
where SGC′ is a real parameter value, y is an output value, and x is an input value;
over the second range by:
y
=
a
′
(
x
Ba
)
2
+
b
′
x
Ba
+
c
′
over the third range by:
y
=
HGC
′
*
x
Ba
+
(
1
-
HGC
′
Ba
)
where HGC′ is a slope; and
where
{
a
′
=
Ba
.
(
-
0.5
·
SGC
′
-
HGC
′
.
MTA
′
)
b
′
=
Ba
-
HGC
′
MTA
′
+
SGC
′
+
HGC
′
2
c
′
=
Ba
-
HGC
′
SGC
′
-
HGC
′
+
MTA
′
2
{
x
SGC
′
=
B
a
-
HGC
′
SGC
′
-
HGC
′
-
MTA
′
2
x
HGC
′
=
B
a
-
HGC
′
SGC
′
-
HGC
′
+
MTA
′
2
where MTA′ is a real parameter value and
{
expgain
=
v
(
L
source
L
target
)
SGC
′
=
expgain
HGC
′
=
highLightGainControl
4
MTA
′
=
midTonesAdjustment
2
with
v
(
x
)
=
log
(
1
+
(
ρ
-
1
)
.
x
1
γ
)
log
(
ρ
)
and
ρ
=
1
+
32.
(
L
target
100000
)
1
/
γ
where midTonesAdjustement is a real value given by:
{
midTonesAdjustment
=
2.21
*
exp
(
-
0.16
*
Ba
)
highlightGainControl
=
CST
where highlightGainControl is a real value, and a modulation value Ba is responsive to a brightness level of the input picture.
13. The method of claim 12 , wherein the real value highlightGainControl is a constant value.
14. The method of claim 12 , wherein the real value highlightGainControl is kept as another parameter in addition to the modulation value.
15. The method of claim 12 , wherein the real value highlightGainControl is given by:
highlightGainControl=2−2*exp(−0.26/ Ba ).
16. The method of claim 1 , wherein the tone-adjustment function is a parametric logarithm function described by:
ba
N
=
Ba
/
PeakL
N
=
log
(
1
+
1
k
.
ba
N
)
y
=
log
(
1
+
x
k
.
ba
N
)
N
,
∀
x
∈
[
0
,
1
]
where k is a constant value, y is an output value, x is an input value, and PeakL is the peak luminance value of the input picture.
17. A method for encoding an input picture comprising:
obtaining a maximum value from component values of the input picture;
obtaining a linear value by tone-mapping the maximum value according to the method of claim 1 ;
multiplying the input picture by a ratio of the linear value over the maximum value.
18. A method for inverse-tone-mapping a tone-mapped version of an input picture by using a parametric inverse-tone-adjustment function described over multiple ranges, at least one parameter of the tone-adjustment function being associated with each range, wherein the method comprises:
determining a non-linear tone adjustment function over at least one range according to a modulation value responsive to the brightness level, and obtaining a parameter of the inverse-tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the input picture.
19. A method for decoding a picture comprising:
obtaining a maximum value from the component values of a decoded picture;
obtaining a non-linear value by inverse-tone-mapping the maximum value according to the method of claim 18 ;
multiplying the decoded picture by a ratio of the non-linear value over the maximum value.
20. A device for tone-mapping an input picture by using a parametric tone-adjustment function described over multiple ranges, at least one parameter of the tone-adjustment function being associated with each range, wherein the device comprises:
a processor configured to determine a non-linear tone adjustment function over at least one range according to a modulation value responsive to the brightness level, and to determine a parameter of the tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the input picture.
21. A device for encoding an input picture comprising a processor configured to:
obtain a maximum value from the component values of the input picture;
obtain a linear value by tone-mapping the maximum value by:
determining a non-linear tone adjustment function over at least one range of a plurality of ranges of a tone-adjustment function according to a modulation value responsive to the brightness level, wherein at least one parameter of the tone-adjustment function is associated with each range of the plurality of ranges; and
determining a parameter of the tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the input picture;
multiply the input picture by a ratio of the linear value over the maximum value.
22. A device for decoding a picture comprising a processor configured to:
obtain a maximum value from the component values of a decoded picture;
obtain a non-linear value by inverse-tone-mapping the maximum value by:
determining a non-linear tone adjustment function over at least one range of a plurality of ranges of an inverse-tone-adjustment function according to a modulation value responsive to the brightness level, wherein at least one parameter of the inverse-tone-adjustment function is associated with each range of the plurality of ranges; and
obtaining a parameter of the inverse-tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the picture;
multiply the decoded picture by a ratio of the non-linear value over the maximum value.
23. A non-transitory computer-readable storage medium comprising instructions that cause a computing device to perform:
determining a non-linear tone adjustment function over at least one range of a plurality of ranges of a tone-adjustment function according to a modulation value responsive to the brightness level, wherein at least one parameter of the tone-adjustment function is associated with each range of the plurality of ranges;
determining a parameter of the tone-adjustment function according to the modulation value, wherein the modulation value is responsive to a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of an input picture; and
tone-mapping the input picture by applying the parametric tone-adjustment function.
24. A device for inverse-tone-mapping a tone-mapped version of an input picture by using a parametric inverse-tone-adjustment function described over multiple ranges, wherein at least one parameter of the inverse-tone-adjustment function is associated with each range, wherein the device comprises a processor configured to determine the inverse-tone-adjustment function as an inverse of a non-linear parametric tone-adjustment function described over at least one range according to a modulation value, wherein a parameter of the non-linear parametric tone-adjustment function is determined according to a modulation value, wherein the modulation value is derived from a mid-tone level of the input picture determined from a black level and a white level, wherein the black level and the white level are based on content of the input picture.