Note: Descriptions are shown in the official language in which they were submitted.
CA 02770759 2012-03-07
METHOD AND SYSTEM FOR CREATING ANIMATION WITH CONTEXTUAL
RIGGING
CROSS-REFERENCE TO RELATED APPLICATIONS
This is the first application filed for the present invention.
TECHNICAL FIELD
The present invention relates to the field of animation creation, and more
particularly
to methods and systems for contextual rigging of animation.
BACKGROUND
In at least some game engines, state machines are used for animating
characters. A
state machine is usually specific to a given character, and comprises
animation clips
which each correspond to a state for the given character.
Control or contextual rigging techniques are also usually used for fixing
animation
clips while game is playing. In some embodiments, a generic control rig is
used for
all of the animation clips. For example, the generic control rig comprises
different
parts for rigging a leg, an arm, a face, etc. Each time the state machine
outputs an
animation clip, an analysis is performed to determine whether each part of the
generic control rig must be applied to the animation clip, which requires
important
computer processing unit (CPU) and memory usage. In other embodiments, several
control rigs each directed to rig a different part of a character are used.
For example,
a first control rig is used for rigging an arm, a second control rig is used
for rigging a
leg, a third control rig is used for rigging a face, etc. Each time the state
machine
outputs an animation clip, an analysis is performed to determine whether each
control rig must be applied to the output animation clip, which also requires
important CPU and memory usage.
- 1 -
CA 02770759 2012-03-07
Therefore, there is a need for an improved method and system for rigging an
animation clip.
SUMMARY
According to a first broad aspect, there is provided a method for applying a
control
rig to an animation of a character, the method comprising: receiving a state
change
for the character being in a first state; determining a second state for the
character
using the state change; retrieving an animation clip and a control rig both
corresponding to the second state, the animation clip comprising a plurality
of poses
for the character each defining a configuration for a body of the character,
the
control rig being specific to the second state and corresponding to at least
one
constraint to be applied on the body of the character; applying the control
rig to the
animation clip, thereby obtaining a rigged animation clip; and outputting the
rigged
animation clip.
In one embodiment, the reception of the state change comprises receiving a
property change for the character.
In one embodiment, the determining step comprises comparing the property
change
to a property threshold.
In one embodiment, the receiving step comprises receiving a desired state for
the
character.
In one embodiment, the application of the control rig to the animation clip
comprises,
for at least some of the plurality of poses, modifying the configuration for
the body of
the character.
In one embodiment, the method further comprises retrieving a control curve
comprising a temporal indication for applying the control rig, said applying
the control
- 2 -
CA 02770759 2012-03-07
rig to the animation clip comprising applying the control rig to each one of
the
plurality of poses according to a corresponding value of the control curve.
In one embodiment, the method further comprises the steps of: receiving a
further
state change for the character; determining a third state for the character
using the
further state change and a state machine; retrieving a further animation clip
for the
character and a further control rig both corresponding to the third state, the
further
animation clip comprising a plurality of second poses for the character, the
further
control rig being specific to the third state and corresponding to at least
one further
constraint to be applied on the body of the character; and blending the
animation clip
and the further animation clip together to obtain a blended animation clip;
wherein
the application of the control rig to the animation clip comprises applying
the control
rig and the further control rig to the blended animation clip.
In one embodiment, the blending step comprises blending at least one last pose
of
the animation clip and at least one first pose of the further animation clip
together to
generate at least one blended pose.
In one embodiment, the control rig is an inverse kinematic rig.
According to a second broad aspect, there is provided an apparatus for
applying a
control rig to an animation of a character, the apparatus comprising a storing
unit
and a processing unit configured for: receiving a state change for the
character
being in a first state; determining a second state for the character using the
state
change; retrieving an animation clip and a control rig both corresponding to
the
second state, the animation clip comprising a plurality of poses for the
character
each defining a configuration for a body of the character, the control rig
being
specific to the second state and corresponding to at least one constraint to
be
applied on the body of the character; applying the control rig to the
animation clip,
thereby obtaining a rigged animation clip; and outputting the rigged animation
clip.
- 3 -
CA 02770759 2012-03-07
According to another broad aspect, there is provided a computer readable
memory
having recorded thereon statements and instructions for execution by a
processing
unit, the statements and instructions comprising the steps of: receiving a
state
change for the character being in a first state; determining a second state
for the
character using the state change; retrieving an animation clip and a control
rig both
corresponding to the second state, the animation clip comprising a plurality
of poses
for the character each defining a configuration for a body of the character,
the
control rig being specific to the second state and corresponding to at least
one
constraint to be applied on the body of the character; applying the control
rig to the
animation clip, thereby obtaining a rigged animation clip; and outputting the
rigged
animation clip.
According to a further broad aspect, there is provided a system for applying a
control
rig to an animation of a character, the system comprising: a state machine
unit
adapted to receive a state change for the character being in a first state,
determine a
second state for the character using the state change, and retrieve an
animation clip
and a control rig both corresponding to the second state, the animation clip
comprising a plurality of poses for the character each defining a
configuration for a
body of the character, the control rig being specific to the second state and
corresponding to at least one constraint to be applied on the body of the
character;
and a rig evaluator adapted to apply the control rig to the animation clip,
thereby
obtaining a rigged animation clip and output the rigged animation clip.
In one embodiment, the state machine unit is adapted to receive a property
change
for the character.
In one embodiment, the state machine unit is adapted to compare the property
change to a property threshold.
In another embodiment, the state machine unit is adapted to receive a desired
state
for the character.
- 4 -
CA 02770759 2012-03-07
In one embodiment, the rig evaluator is adapted to modify the configuration
for the
body of the character for at least one of the plurality of poses.
In one embodiment, the state machine unit is further adapted to retrieve a
control
curve comprising a temporal indication for applying the control rig and the
rig
evaluator is further adapted to apply the control rig to each one of the
plurality of
poses according to a corresponding value of the control curve.
In one embodiment, the system further comprises a blending unit connected
between the state machine unit and the rig evaluator, the state machine unit
being
further adapted to receive a further state change for the character, determine
a third
state for the character using the further state change, and retrieve a further
animation clip for the character and a further control rig both corresponding
to the
third state, the further animation clip comprising a plurality of second poses
for the
character, the further control rig being specific to the third state and
corresponding to
at least one further constraint to be applied on the body of the character,
the
blending unit being adapted to blend the animation clip and the further
animation clip
together to obtain a blended animation clip, and the rig evaluator being
further
adapted to apply the control rig and the further control rig to the blended
animation
clip.
In one embodiment, the blending unit is adapted to blend at least one last
pose of
the animation clip and at least one first pose of the further animation clip
together to
generate at least one blended pose.
In one embodiment, the control rig is an inverse kinematic rig.
It should be understood that a "character" can be any entity that can be
animated in
an animation, a video game, and the like. For example, a character can be an
animated object, an animated avatar, or the like.
- 5 -
CA 02770759 2012-03-07
BRIEF DESCRIPTION OF THE DRAWINGS
Further features and advantages of the present invention will become apparent
from
the following detailed description, taken in combination with the appended
drawings,
in which:
Fig. 1 is a block diagram illustrating a state machine, in accordance with an
embodiment;
Fig. 2 illustrates an exemplary pose for a character, in accordance with an
embodiment;
Figs. 3a-3c illustrate three poses comprised in an animation clip
corresponding to a
"walk" state, in accordance with an embodiment;
Fig. 4 is a flow chart of a method for applying a control rig to an animation
clip, in
accordance with an embodiment;
Figs. 5a-5f illustrate different poses comprised in an animation clip
corresponding to
a "jump" state, in accordance with an embodiment;
Figs. 6a-6f illustrate different poses comprised in animation clips
corresponding to a
"grab gun" and "aim at" states, in accordance with an embodiment;
Fig. 7 is a block diagram illustrating a system for applying a control rig to
an
animation clip, in accordance with an embodiment; and
Fig. 8 is a block diagram illustrating a game engine, in accordance with an
embodiment.
It will be noted that throughout the appended drawings, like features are
identified by
like reference numerals.
- 6 -
CA 02770759 2012-03-07
DETAILED DESCRIPTION
In the present description, there is described a method and system for
applying a
control or contextual rig to an animation. The animation is created using a
state
machine. A state machine comprises a network of states which represents the
different states in which an animated character or object may be. The
different states
are interconnected by connections which each represent a transition between
two
interconnected states. A state-specific animation clip is associated with each
state of
the state machine.
At least one state-specific control or contextual rig is associated with at
least one of
the states of the state machine. When playing an animation, only the control
rig
associated with the state corresponding to the animation is evaluated to
determine
whether the control rig should be applied. Therefore, if no control rig is
associated
with a given state, no evaluation is required when the animation clip
corresponding
to the given state is played, which reduces the computational time and/or the
memory requirements.
A control or contextual rig comprises at least one constraint which performs a
specific task on at least one part of the body of an animated entity, such as
an
animated character or animated object for example. For example, a control rig
can
comprise a chain solver. A control rig may comprise constraints to force a
character
to look at a particular object or in a particular direction. In another
example, a control
rig may comprise constraints to impose a given position/configuration for the
animated character or for a part of the body of the animated character. In a
further
example, a control rig may be used to ensure a physical contact of the foot of
a
character on top of the ground, and keep the character knee pointing in a
desired
direction.
When applying the control rig to an animation clip, the configuration of the
character
body is modified so that the constraints corresponding to the control rig be
met.
- 7 -
CA 02770759 2012-03-07
In one embodiment, during a transition between a first state and a second
state for
the animated character, the animations corresponding to the first and second
states
are first blended together before being rigged and played. Only the control
rigs
associated with the first and second states are successively evaluated after
the
The animation generation for a character is based on a state machine. As
described
above, a state machine comprises a network of possible states in which an
animated
character may be. Figure 1 illustrates one exemplary network of states for a
character. The central state of the network is the "idle" state. The other
possible
15 the first and second states are connected together. If two states are
directly
connected together, the character can pass directly from one of these two
states to
the other. For example, the character can directly pass from "idle" to "turn-
left" or
from "walk" to "run". However, if a first and a second state are indirectly
connected
together, i.e. connected via at least one intermediary state, then the
character
20 cannot directly pass from the first state to the second state, and vice
versa. The
character has to pass through the at least one intermediary step to pass from
the
first state to the second state, and vice versa. For example, while he can
directly
pass from the "run" state to the "walk" state, the character has to pass
through the
"stance-to-walk" and "step front" states to pass from the "walk" state to the
"idle"
A state-specific animation clip is associated with each state comprised in the
state
machine. Each animation clip comprises information that can be used for
generating
an animation of a character when in the state to which the animation clip is
- 8 -
CA 02770759 2012-03-07
associated. For example, the animation clip associated with a "walk" state
comprises
information from which an animation in which a character is walking may be
generated. For example, an animation clip comprises at least a series of
poses,
each defining a given configuration for the body of the character. An
animation clip
may also comprise a sample rate, i.e. the sample rate at which the animation
clip
was created.
Figure 2 illustrates one exemplary pose for a character. The pose corresponds
to a
given configuration for the body of the character to which the animation clip
is
associated. For example, a body configuration may be defined by a position, a
rotation, and a scale for the skeleton joints and/or bones of the character
body.
In one embodiment, the joints and/or bones of the character body are linked to
control points that can be used to control the orientation of body parts of
the
character.
As described above, an animation clip comprises a series of poses of which a
timely
execution creates an animation. Figures 3a, 3b, and 3c illustrates some poses
comprises in an animation clip corresponding to a "walk" state.
Figure 4 illustrates one embodiment of a method 40 for applying a control rig
to an
animation of a character during execution of a video game. The animation is
based
on a state machine specific to the character to be animated. The first step 42
of the
method 40 comprises the step of receiving a change for the state of the
character
that is in a present or first state. A state change comprises any input that
may affect
the state in which the character is and force the character to pass from the
first state
to a second state.
In one embodiment, the state change comprises an identification of a desired
state
for the character. For example and referring back to Figure 1, the received
state
change may be indicative of the "step back" state while the character is
presently in
- 9 -
CA 02770759 2012-03-07
the "idle" state. In another example, the received state change may be
indicative of
the "idle" state while the character is presently in the "run" state.
In another embodiment, the state change may be indicative of a change in a
characteristic or parameter of the character. For example, the character may
be
presently in a "run" state and, at step 42, a change in the speed at which the
character is running is received.
At step 44, the second state for the character is determined using the
received input,
i.e. the state change, and the state machine, such as the state machine
illustrated in
Figure 1.
In an embodiment in which the received state change information is indicative
of a
desired state for the character, a determination of the interconnection
between the
present or first state and the desired state is performed. If the first state
and the
desired state are directly connected together, then the second state is the
desired
state comprised in the received state change. For example, if the character is
presently in the "idle" state and the desired state indicated in the received
state
change is the "step back" state, then the second state is the "step back"
state since
the "idle" and "step back" states are directly connected. If the first state
and the
desired states are not directly connected together, i.e. if they are connected
via at
least one intermediary state, then the second state is the first intermediary
state. For
example, if the character is presently in the "run" state and the desired
state
indicated in the received state change is the "idle" state, then the "run" and
"idle"
states are indirectly connected via three intermediary states, i.e. the
"walk", "stance-
to-walk", and "step front" states. In that case, the second state is the first
intermediary state, i.e. the "walk" state.
In an embodiment in which the received state change information is indicative
of a
character parameter, the second state is determined using the character
parameter.
For example, the value of the parameter may be compared to a threshold and the
-10-
CA 02770759 2012-03-07
second state is determined according to the result of the comparison. For
example,
a parameter chosen for determining the second state may be the speed of a
character being in a "walk" state. If the speed of the character decreases
below a
first threshold, then it is determined that second state is the "idle" state.
If the speed
of the character increases above a second threshold being greater than the
first
threshold, then it is determined that the second state for the character is
the "run"
state.
At step 46, both an animation clip and a control rig, each specific to the
second
state, are retrieved from a database. The database comprises a state-specific
animation clip for each state of the state machine. The database further
comprises a
state-specific control rig for at least some of the states comprised in the
state
machine.
It should be understood that if the second state is associated with no control
rig, then
only the animation clip specific to the second state is retrieved. In that
case, the
animation clip is then output without any control rigging step.
At step 48, a determination as so whether the state-specific control rig
should be
applied to the animation clip is performed. If the state-specific control rig
should not
be applied to the animation clip, then the animation clip is output without
any control
rigging. Otherwise, the state-specific control rig associated with the second
state is
applied to the animation clip corresponding to the second state. When applying
the
control rig to the animation clip/pose, the constraints comprised in the
control rig are
applied to the character, i.e. the configuration of the character is modified
so that the
constraints be met. For example, the position of a character's foot in a pose
may be
modified to ensure a physical contact between the character's foot and a
floor.
In one embodiment, for each pose of the animation clip, a determination as to
whether the control rig should be applied is performed. Depending on the
result of
-11 -
CA 02770759 2012-03-07
the determination, the control rig may be applied to each pose of the
animation clip
or only to some of the poses of the animation clip.
In one embodiment, the application of the control rig to the animation
clip/poses is
performed using information contained in the animation clip. For example, in
an
animation clip in which a character wears a holster for receiving a handgun, a
control
rig may comprise constraints to ensure that the character's hand will be in
physical
contact with the handgun while the character draws the handgun. In this case,
the
position of the handgun, while in the holster, is comprised in the animation
clip.
In the same or another embodiment, the application of the control rig is
performed
using information external to the animation clip, such as timing, the position
of an
object, with which the character has to interact during the animation, the
position for
a body part of the character or information from which the position for the
body part
can be determined, the height of an obstacle to jump over, the current slope
of a
terrain on which the character walks, the positions of the hands on a edge
when
climbing a wall, and the like. For example, if in an animation clip a
character has to
jump on an object, a control rig may be used for ensuring that the character's
hand
will grab the top end of the object while jumping.
In one embodiment, a control curve is further associated with an animation
clip and
a control rig. The control curve provides a temporal indication for the
execution of
the control rig and the application of the constraints on the character. For
example
and referring to the previous example, the control curve may temporally
indicate
when the character's hand should grab the top end of the object. In one
embodiment, for each pose of an animation clip, the control curve indicates
whether
the control rig should be applied or not. In another embodiment, the control
curve
provides a percentage for each pose. For example, for a control rig value of
25%,
the control rig is first applied to the animation clip/pose in order to obtain
a first
rigged animation clip/pose. Then, the first rigged animation clip/pose is
blended with
the original animation clip/pose to obtain a second rigged animation clip/pose
which
- 12 -
CA 02770759 2012-03-07
corresponds to 75% of the original animation clip/pose and 25% of the first
rigged
animation clip/pose.
Once the control rig has been applied to the animation clip, the rigged
animation clip
is sent to a rendering unit which generates an animation from the poses of the
animation clip. The animation is then stored on memory and/or sent to a
display unit.
In one embodiment, at least the last pose of the first animation clip and the
first pose
of the second animation clip are blended together in order to create a
transition
between the first and second animation clips. The poses are first blended
together
before applying the control rig(s) to the blended pose. In this case, the
control rigs
associated with the first and second animation clips are successively applied
to the
blended pose.
Figures 5a-5f illustrates one example in which a control rig is applied to an
animation
clip using a control curve. In the animation clip, a character jumps on top of
a cube.
The control rig associated to the animation clip comprises constraints to
ensure that
the hands of the character will be in physical contact with the top edge of
the cube
during the jump. The control curve associated with the animation clip and the
control
rig provides timing for the application of the control rig to the animation
clip.
Figure 5a illustrates a pose in which the character is in an "idle" state
facing the cube
on top of which he has to jump. A state change indicative of a "jump" state is
then
received, for example. The animation clip and the control rig both specific to
the new
state "jump" are then retrieved. The control curve associated with the control
rig and
the animation clip is further retrieved.
For each pose of the animation clip, it is determined whether the control rig
should
be applied using the control curve. If the control rig has to be applied to
the pose, the
constraints corresponding to the control rig are applied to the character and
the
character configuration is modified so that the character configuration meets
the
constraints. In the present example, the position of the hands of the
character is
- 13-
CA 02770759 2012-03-07
modified using a target position for the hands of the character on the cube
and the
current position of the hands of the character within the pose to ensure that
the
character grabs the top edge of the cube.
Figure 5b illustrates the 15th pose of the animation clip in which the
character is
preparing to jump. The control curve value associated with the 15th pose is
equal to
zero. Therefore, no control rig should be applied to the 15th pose illustrated
in Figure
5b. Similarly, Figure 5c illustrates the 20th pose of the animation clip in
which the
character is jumping. Since the control curve value for the 20th pose is equal
to zero,
the control rig is not applied.
Figure 5d illustrates the 22"d pose of the animation clip in which the hands
of the
character are reaching the top edge of the cube. The control curve value
associated
with the 22nd pose is equal to 100%. Therefore, the control rig is applied to
the 22nd
pose. The constraints comprised in the control rig are applied to the
character and
the configuration of the character is modified for ensuring that the hands of
the
character are in physical contact with the top edge of the cube.
Figures 5e and 5f illustrate the 40th and 55th poses of the animation clip,
respectively, in which the character is progressively standing up on top of
the cube.
The control curve value associated with the two poses is equal to zero.
Therefore,
no control rig is applied to the 40th and 55th poses.
Figures 6a-6e illustrate a further example in which two control rigs are
applied to two
animation clips using two control curves, each for a respective control rig.
In the first
animation clip, a character draws a handgun from a holster while in the second
animation clip, the character aims at a given direction/point. A first control
rig
associated to the first animation clip comprises constraints to ensure that
the hand of
the character will be in physical contact with the handgun while the character
draws
the handgun. A second control rig associated to the second animation clip
comprises constraints to ensure that the handgun aims at the given
direction/point. A
- 14 -
CA 02770759 2012-03-07
first and a second control curves associated with the first and second control
rigs,
respectively, provides timing for the application of the first and second
control rigs to
the first and second animation clips, respectively.
Figure 6a illustrates a pose in which the character is in an "idle" state. A
state
change indicative of a "handgun draw" or "grab" state is then received, for
example.
The first animation clip and the first control rig both specific to the new
state
"handgun draw" are then retrieved. The first control curve associated with the
first
control rig and the first animation clip are further retrieved.
For each pose of the first animation clip, it is determined whether the first
control rig
should be applied using the first control curve. If the first control rig
should be
applied to the pose, the constraints corresponding to the first control rig
are applied
to the character and the character configuration is modified so that the
character
configuration meets the constraints. In the present example, the position of
the arm
and hand of the character is modified using the position of the handgun or
holster
(not shown) and the current position of the hand of the character to ensure
that the
character grabs the handgun.
Figure 6b illustrates a pose of the first animation clip before the hand of
the
character grabs the handgun. The first control curve value associated with the
pose
is equal to zero. Therefore, the first control rig is not applied to the pose
illustrated in
Figure 6b. Figure 6c illustrates a pose of the animation clip in which the
character
partially grabs the handgun just before firmly holding the handgun. The first
control
curve value for pose is equal to 55%. In this case, the resulting rigged pose
corresponds to 45% of the original or non-rigged pose and 55% of a pose to
which
the control rig has been fully applied. Therefore, the first control rig is
partially
applied to the pose.
Figure 6d illustrates a pose of the animation clip in which the character
firmly holds
the handgun. The first control curve value associated with the pose is equal
to
-15-
CA 02770759 2012-03-07
100%. Therefore, the first control rig is fully applied to the pose. The
constraints
comprised in the first control rig are applied to the character and the
configuration of
the character is modified for ensuring that the character firmly holds the
handgun in
his hand.
A state change indicative of an "aim-at" state is then received, for example.
The
second animation clip and the second control rig both specific to the new
state "aim-
at" are then retrieved. The second control curve associated with the second
control
rig and the second animation clip is further retrieved.
The second animation clip comprises information for generating an animation in
which a character holding an object in his hand aims at a given direction or
point.
For each pose of the second animation clip, it is determined whether the
second
control rig should be applied using the second control curve. If the second
control rig
should be applied to the pose, the constraints corresponding to the second
control
rig are applied to the character and the character configuration is modified
so that
the character configuration meets the constraints. In the present example, the
position of the arm and hand of the character is modified using a target
direction/point and the current position of the hand/arm of the character to
ensure
that the handgun aims at the target direction/point.
Figure 6e illustrates a pose of the second animation clip in which the
character aims
the handgun. The second control curve value associated with the pose
illustrated in
Figure 6e is equal to 100%. Therefore, the second control rig is applied to
the pose
and the configuration of the character is modified to ensure that the handgun
aims at
the target direction.
While the above description refers to a method 40 for applying a control or
contextual rig to an animation clip, it should be understood that the method
40 can
be embodied as an apparatus comprising a processing unit and a memory. The
processing unit is configured for performing the steps of the method 40.
- 16 -
CA 02770759 2012-03-07
In one embodiment, the method 40 is embodied as a computer readable memory
having recorded thereon statements and instructions for execution by a
processing
unit. The statements and instructions comprise the steps of the method 40.
Figure 7 illustrates a rigging system 70 for applying a control or contextual
rig to an
animation of a character during execution of a video game or animation
creation.
The system 70 comprises a state machine unit 72 and a rig evaluator 74.
The state machine unit 72 comprises a state machine specific to the character
to be
animated and a database of animation clips and control rigs. Each animation
clip
and each control rig are associated to a corresponding state of the state
machine.
Each state is associated with a corresponding animation clip while only some
states
may be associated with a corresponding control rig. Alternatively, all of the
states
may be associated with a corresponding control rig.
The state machine unit 72 is adapted to receive a change for the state of the
character that is in a present or first state. As described above, a state
change
comprises any input that may affect the state in which the character is and
force the
character to pass from the first state to a second state. For example, the
state
change may comprise an identification of a desired second state for the
character. In
another example, the state change may be indicative of a change in a
characteristic
or parameter of the character from which the second can be determined.
The state machine unit 72 is further adapted to determine the next or second
state
for the character using the received state change and the state machine, as
described above.
In an embodiment in which the received state change is indicative of a desired
state
for the character, the state machine unit is adapted to determine the
interconnection
between the present or first state and the desired state using the state
machine. If
the first state and the desired state are directly connected together, then
the second
state is the desired state comprised in the received state change. If the
first state
-17-
CA 02770759 2012-03-07
and the desired states are not directly connected together, i.e. if they are
connected
via at least one intermediary state, then the second state is the first
intermediary
state.
In an embodiment in which the received state change information is indicative
of a
character parameter, the state machine unit 72 is adapted to determine the
second
state using the character parameter. For example, the value of the parameter
may
be compared to a threshold and the second state is determined according to the
result of the comparison. For example, a parameter chosen for determining the
second state may be the speed of a character being in a "walk" state. If the
speed of
the character decreases below a first threshold, then it is determined that
second
state is the "idle" state. If the speed of the character increases above a
second
threshold being greater than the first threshold, then it is determined that
the second
state for the character is the "run" state.
The state machine unit 72 is further adapted to retrieve an animation clip and
a
control rig, each specific to the second state, from the database. The state
machine
unit 72 outputs the retrieved animation clip and the corresponding control
rig. In one
embodiment, the state machine unit 72 is adapted to successively output each
pose
of the retrieved animation clip.
It should be understood that if the second state is associated with no control
rig, then
the state machine unit 72 only retrieves the animation clip specific to the
second
state, and the rig evaluator 74 outputs the animation clip without applying
any control
rigging to the animation clip.
The rig evaluator 74 is further adapted to receive the animation clip and the
control
rig from the state machine unit 72, and determine whether the state-specific
control
rig should be applied to the animation clip. If the state-specific control rig
should not
be applied to the animation clip, then the rig evaluator 74 outputs the
animation clip
without any control rigging. Otherwise, the rig evaluator 74 applies the state-
specific
- 18 -
CA 02770759 2012-03-07
control rig associated with the second state to the animation clip
corresponding to
the second state. When applying the control rig to the animation clip/pose,
the
constraints comprised in the control rig are applied to the character, i.e.
the
configuration of the character is modified so that the constraints be met. For
example, the position/configuration of a character's foot and leg in a pose
may be
modified to ensure a physical contact between the character foot and a floor.
In an embodiment in which the state machine unit 72 outputs the animation clip
pose
by pose at a given frame rate, the rig evaluator 74 is adapted to determine
whether
the control rig should be applied to each pose of the animation clip.
Depending on
the result of the determination, the control rig may be applied to each pose
of the
animation clip, some of the poses of the animation clip, or only one pose.
In one embodiment, the rig evaluator 74 uses information contained in the
animation
clip to apply the control rig to the animation clip/pose. For example, in an
animation
clip in which a character wears a holster for receiving a handgun, a control
rig may
comprise constraints to ensure that the character's hand will be in physical
contact
with the handgun while the character draws the handgun. In this case, the
position of
the handgun, while in the holster, is comprised in the animation clip and used
by the
rig evaluator 74 to ensure that the hand of the character will firmly hold the
handgun.
In the same or another embodiment, the rig evaluator 74 is adapted to apply
the
control rig using information external to the animation clip, such as video
game
timing, the position of an object, with which the character has to interact
during the
animation, the position for a body part of the character or information from
which the
position for the body part can be determined, and the like. For example, if in
an
animation clip a character has to jump on an object, a control rig may be used
for
ensuring that the character's hand will grab the top end of the object while
jumping.
In one embodiment, the database of the state machine unit 72 further comprises
a
control curve associated with an animation clip and a control rig. As
described
-19-
CA 02770759 2012-03-07
above, the control curve provides a temporal indication for the execution of
the
control rig and the application of the constraints on the character. For
example and
referring to the previous example, the control curve may temporally indicate
when
the character's hand should grab the top end of the object. The state machine
unit
72 is further adapted to output the control curve along with the animation
clip and the
control rig to the rig evaluator 74 which uses the control curve to determine
whether
and/or when the control rig should be applied.
In an embodiment in which it outputs the poses of the animation clip
separately and
successively, for each pose of an animation clip, the state machine unit 72
outputs
the control value corresponding to each pose along with the pose. For example,
the
control curve value may be represented by a percentage for each pose as
described
above.
Once the rig evaluator 74 has applied the control rig to the animation clip,
the rigged
animation clip is sent to a rendering unit which generates an animation from
the
poses of the animation clip. The animation is then stored on memory and/or
sent to
a display unit.
In one embodiment, the rigging system 70 further comprises a blending unit
(not
shown) for blending at least the last pose of the first animation clip and the
first pose
of the second animation clip together in order to create a transition between
the first
and second animation clips. The blending unit is located between the state
machine
unit 72 and the rig evaluator 74. In this case, the poses are first blended
together by
the blending unit before application of the control rig(s) to the blended pose
by the
rig evaluator 74. In this case, the control rigs associated with the first and
second
animation clips are successively applied to the blended pose which is sent to
the rig
evaluator by the blending unit.
Figure 8 illustrates one embodiment of a game engine 80 for the creation
and/or
development of video games. The game engine 80 comprises a sounds engine unit
- 20 -
CA 02770759 2012-03-07
82, a graphics engine unit 84, a physics engine unit 86, a game logic or
artificial
intelligence unit 88, and an animation engine unit 90 which comprises the
state
machine unit 72, the rig evaluator 74, and a rendering unit 92 for rendering
the
poses output by the rig evaluator 74. The rendered poses are then displayed on
a
display unit (not shown).
The sounds engine unit 82 is adapted to provide playback of interactive audio
files of
diverse formats. The graphics engine unit is adapted to generate images from a
scene. In one embodiment, a scene contains objects that contain geometry,
viewpoint, texture, lightning, and shading information. The physics engine
unit 86 is
adapted to provide an approximate simulation of certain physical systems, such
as
rigid body dynamics (including collision detection), soft body dynamics, and
fluid
dynamics, etc.
It should be understood that the game engine 80 may comprise more or less
modules /units as long as it comprises the state machine unit 72 and the rig
evaluator 74.
In one embodiment, the control or contextual rig is an inverse kinematic (IK)
rig. In
another embodiment, the control or contextual rig is a forward kinematic (FK)
rig.
While the present description refers to a method and system for applying a
control or
contextual rig to an animation in the context of a video game, it should be
understood that the method and system can be used in the context of animation
creation.
The embodiments described above are intended to be exemplary only. The scope
of
the invention is therefore intended to be limited solely by the scope of the
appended
claims.
- 21 -
