Posted by Michael on June 17th, 2006, in Development

Perhaps underacting is a form of stylisation that can help us limit the amount of hand-crafted character animations that we need. For this purpose, I studied Humphrey Bogart’s and Lauren Bacall’s acting in Key Largo with the sound off.

Here’s some observations that don’t necessarily apply to underacting as such but perhaps to human behaviour in general.


A very inspiring thing regarding idle behaviour -which came up a few times before in this journal- was that I noticed that an real actor doesn’t have a neutral pose. He does a certain motion and then freezes in the end pose of this movement. Later he does another motion that ends in another pose.

This could be translated to virtual actors if, rather than blending out of and into a neutral stance in between actions (which most games, including The Endless Forest, default to), at the end of an action, an idle animation is played that pauses in a certain pose (with a loop for subtle motion so the characters don’t look frozen). Then when another action is chosen, the idle motion continues to play to its end while the action animation blends in. In theory, there would still be a neutral stance at the end of the idle motion and at the beginning of the action, but it would disappear in the transition, or at least be displayed only for a very short time.

Walking speed is used for expression: slowing down means attention, speeding up means enthusiasm or urgency.

In 8 and in The Endless Forest, the character walks fast if the target is far away and slow if it is closeby, thus ignoring the expressive potential of walking speed.

When something happens, e.g. when someone talks to the actor, the actor’s body does not respond immediately. He sits there motionless and we think he’s listening. Only a little later does he turn his head towards the event.

When a third person arrives, the actor looks at her and steps back a little to make room for her in the circle of the conversation.

When the actor looks down briefly during an interaction, it expresses thought or (mild) embarassment.

When an actor stays in the same pose while another is talking, he seems to be listening, even if he doesn’t look at the other. In fact, not looking seems like he’s listening more intently.

When several actors are in a scene, the viewer pays attention to the one who does the action, more or less ignoring all the others.

Bogart almost never holds his head straight. It’s always a little cocked. At least when he’s sitting or standing still.

When a woman touches her hair in the company of a man, it expresses romantic interest.

An actor can be standing still until he is spoken to. Then he moves into another pose (hands in back, e.g.).

When a shock happens, the actor changes pose as well.

When the actor is still in a pose and wants to move, he will first look elsewhere.

The actor stands still when listening and wobbles when talking, especially when starting to talk.

Just staring at somebody without moving and without facial expression expresses hatred. When the eyes are wide open, it expresses aggression.

Unrelated to actors but interesting nonetheless: a slow ventilator expresses heat. 🙂

No Comments »

Posted by Michael on June 16th, 2006, in Development

The core of Drama Princess is playing canned character animations. This is a very artificial situation. Human actors put a lot more variation in their performance. They never really do the same things more than once, certainly not within the course of a single play. Virtual actors, with the current state of technology, are not that versatile.

The risk exists that, despite our efforts, our Drama Princess actor will still seem artificial and robot-like. We could compensate to some extent by making large amounts of animations for each character. But that costs a lot of production money, takes a lot of production time and increases the file size and probably the memory requirements as well. And ultimately, given the nature of the system, the actor will repeat its motions once in a while anyway.

It looks like repetitive behaviour is a fact of life when working with virtual actors. The question now becomes, how do we work with this limitation? What can we do within this limitation to maximize believability.

One solution could perhaps be found in stylisation. Certain forms of theater have embraced such stylisation. The Commedia dell’Arte is a well known example that inspired modern artists like Meyerhold to develop an extremely stylised form of acting. Since the vocabulary of such acting is limited, it could serve as a model for a virtual actor system.

The more expressive animations are, the more you will notice when they repeat. If animations are subtle, however, you will be far less likely to notice the repetition.

Perhaps the underacting that was typical for Hollywood films in the 40s and 50s could be inspiring as well. Humphrey Bogart and Lauren Bacall are masters in acting out passionate feelings while hardly moving a limb or batting a lid. We’ll have to study their work a bit to find out if they are really immobile or if there is still a lot going on that is possibly far too subtle for any virtual actor.

3 Comments »

Posted by Michael on June 16th, 2006, in Development

I made a first little text-based test with a very limited version of the Romanicism idea.

Some observations:

Despite the randomness, the behaviour in the beginning is (almost?) always the same. When shocks start happening, more variation occurs. So perhaps we shouldn’t wait until a relationship gets saturated to generate a shock. Perhaps mild shocks as the relationship develops, would appropriate.
Distributed “mini-shocks” would also allow the actor to express its feelings once in a while.

At some points, the girl seems to cycle through the same set of behaviours. So we may want to add a bit of randomness to the restoration of fascination.

The girl is always doing something. This will probably make her look unrealistic or nervous. In the romanticism concept, this would be solved by adding more opportunities to the environment (in this case, the room). But that would still make her choose between the ball and the room and might still make her interact with the ball too much. Maybe we should introduce a new element: idle behaviour. Just standing around, doing nothing much, looking at objects, or staring in space. This could be influenced by a variable like comfort. The more comfortable an actor feels, the less it will be inclined to choose an opportunity as a goal and the more it will choose to do idle behaviour. This could be used to help define the actor’s personality (fixed) or its mood (dynamic).

1 Comment »

Posted by Michael on June 14th, 2006, in Development

This is basically a new version of Consumerism, named after a suggestion by Patrick. I’ll try to be brief for the sake of clarity. The details can be dealt with in seperate posts.

Drama Princess is by no means an attempt to recreate human life. Its only purpose is to model autonomous behaviour of characters in such a way that the user can imagine things about it. Rather than realistic or even believable, we want our characters to be suggestive and inspiring.

A character that interacts with something is called an actor. The (potential) target of such an interaction is called object. An object can be another character, an inanimate item or an environment.

Intimacy is the level of the relationship of an actor with each object in the world. The actor starts the game with intimacy values for categories of objects only. When it interacts with an individual, it creates an intimacy value for this object. Each interaction with an object changes this intimacy value.

Enthusiasm defines attention span and the eagerness with which an actor will try to improve its intimacy with an object. A very enthusiastic actor will switch more easily between different objects to interact with. Enthusiasm remains the same throughout the game (unless aging is part of it).

The author uses intimacy and enthusiasm to define the personality of each actor.

All objects in the game world have a list of interactions that actors can perform with them, called opportunities. Each actor continuously collects all opportunities in its vicinity and chooses one of them as its goal. It will attempt to achieve this goal by performing a sequence of actions.

All activity in the game is interactivity with objects.
Solitary activity is defined as interactivity with the environment.

Every opportunity fits with a certain level of intimacy. An actor will select an opportunity as its goal based on its level of intimacy with the object that broadcasted the opportunity. Enthusiastic actors will be more inclined to choose an opportunity that is more intimate than fits their relationship with the object. Less enthusiastic actors will choose one the fits more closely.

Every time a goal is achieved, the actor’s intimacy with the object that broadcasted the opportunity is altered. Most of the time, the intimacy will improve (the reasoning being that interaction with objects increases familiarity with them). But the bigger the difference between the level of intimacy of the opportunity and the level of intimacy of the relationship, the higher the chance of a negative effect (because the other character does not want to be rushed when you do something that is too enthusiastic or because it is disappointed when you do something that seems too distant).

When an actor is seperated from an object, its intimacy with this object goes down over time. Intimacy is a value that is shared by both participants in the relationship (for the sake of simplicity, they are always equally fond of each other).

When intimacy with an object has exceeded its maximum value, the chance of a break-up will grow. When this happens, the intimacy between the two will be reset to a lower level.

When intimacy drops below zero, actors will be more inclined to choose opportunities that are even lower (i.e. you want to reduce the intimacy with your enemies).

Since interactions alter the intimacy, the preference of an actor for opportunities will continuously change. To completely exclude the chance of an actor selecting the same opportunity twice in a row, fascination for the opportunity drops instantly after the goal has been achieved. Over time, fascination reaches a normal level again.
To ensure that actors don’t keep pursuing goals that they cannot achieve, fascination of the opportunity that has been chosen as the goal decreases over time. At some point other opportunities will seem more appealing and the actor will choose a new goal.

The concept of shock ensures that characters respond to sudden events. A shock will replace the global list of opportunities with a limited list of responses to the event, out of which the characters can choose.

8 Comments »

Posted by Michael on June 13th, 2006, in Development

The Oz papers tell me that expressing emotions is very important for the characters to be believable (which is something they in turn learned from the Disney animators). And my Body Language book tells me that when people like each other, they start imitating each other’s body language, at least in the early stage of the friendship.

So it seems that it would help believabilty if characters would express the effect of an interaction with each other through body language (rather than by having plus or minus signs float above their heads as in The Sims). So that when the interaction (i.e. “achieving the goal” in Consumerism speak) results in a positive effect on the relationship, the characters briefly imitate each other’s posture. Or when the effect was negative, they purposely don’t.
Perhaps the animation sequence that is played during such an interaction can end in a certain pose (it has to go back to some neutral stance anyway) which is an invitation for the other character to either imitate or ignore.

Even if this would look artificial, it could still be powerful. I only wonder whether this can be read by the player (either consciously or subconsciously). I know of only one other game that uses imitation to express relationship: in Animal Crossing a character changes into the other character’s clothing to express fondess. This is very clear to the player.

11 Comments »

Posted by Michael on June 13th, 2006, in Books

read in a Dutch translation by Saskia Tijsma

This is sort of a self-help book that teaches how to read and produce body language in order to be socially succesful. Its subtitle is “a manual for non-verbal communication”. So it’s not very academic and most of what’s in the book is very obvious (“women’s magazine style” 😉 ). But here and there, a few interesting things are mentioned.

1 Comment »

Posted by Michael on June 13th, 2006, in Projects

These are quotes from papers written in the context of the Oz research project.

I’m still reading these texts, so this post may be updated later.

Form “The Role of Emotion in Believable Agents” by Joseph Bates (April 1994)

“It can be argued that while scientists may have more effectively recreated scientists, it is the artists who have come closest to understanding and perhaps capturing the essence of humanity (…)”
(referring to A.I. scientists and Disney animators)

“The apparent desires of a character, and the way the character feels about what happens in the world with respect to those desires, are what makes us care about that character. If the character does not react emotionally to events, if they don’t care, then neither will we.”

“Jones says he discovered that it is the oddity, the quirk, that gives personality to a character, and it is personality that gives life.”
(referring to animator of Bugs Bunny and a programming error in one of the Woggles)

From “Believable Social and Emotional Agents” by W. Scott Neal Reilly (May 1996)

“My framework gets a lot of its power from being part of a broad agent architecture. The concept is simple: the agent will be emotionally richer if there are more things to have emotions about and more ways to express them. This reliance on breadth has also meant that I have been able to create simple emotion models that rely on perception and motivation instead of deep modeling of other agents and complex cognitive processing.”

“The goal of building believable agents is inherently an artistic one. Traditional AI goals of creating competence and building models of human cognition are only tangentially related because creating believability is not the same as creating intelligence or realism.”

“From the standpoint of believability, it is better to go with the less realistic characters which meet the audience’s expectations than to go with the more realistic characters which don’t.”

2 Comments »

Posted by Michael on June 10th, 2006, in Development

All objects (including characters) in the game world advertise their opportunities. Mostly based on its relationship with these objects, a Drama Princess character chooses one of these opportunities as its goal.
An opportunity consists of several animations played in sequence. When a character chooses one as its goal, it’s going to play these animations one after the other. Some sequences will simply play from start to finish, after which a new goal can be chosen. Others may pause by putting the character in a certain state, typically expressed by a looping animation. After a while the character will get bored with this opportunity and it will finish the sequence by playing the final animation.

When a sequence has finished and before the character chooses a new goal, the character is in an idle state. To have the character return to the same kind of idle behaviour, like a neutral stance (as happens in The Sims and The Endless Forest) , looks very unnatural. To solve this issue our environment objects will offer several idle behaviours that are available for the current environment. So rather than going to a neutral stance when an animation sequence ends, the character will select an idle behaviour from the ones offered by the environment.

When a character is moving towards a relatively uninteresting goal and meets another character with which it has a good relationship, there’s a very good chance that it will interrupt what it was doing in favour of interacting with the other character. So rather than finishing the sequence and selecting an idle behaviour from the environment object, it will immediately start playing the animation sequence of the opportunity offered by the other character that it chose as its new goal.
For both cases, idle behaviour and interruption, a transition between two animation sequences will be required. Since so many combinations are possible (especially in the case of interruption), we cannot hard-code all transition animations.

Blending between two random poses almost always looks unnatural and can in some cases produce unwanted effects (like character rotating through the floor).

But we could simplify the problem by stylizing of our concept of invitation. When a character chooses to do an interaction with another character, it will send an invitation to this character and wait for it to respond and move to the right location and into the right starting pose. Then the interaction can happen, nice and synchronized. This moment of invitation could be made explicit in a certain pose that is always the same. Then the complexity of transitioning is at least reduced by half: we know the end of the transition.

The problem is that this may end up looking like a neutral stance in between animation sequences, mentioned above as decidedly unnatural. This is hardly acceptable. But it may be an area where we need to compromise. Perhaps extreme stylization (maybe in combination with camera cuts) may help make this more acceptable.

3 Comments »

Posted by Michael on June 10th, 2006, in Development

This is a refinement of the decision making process suggested in Consumerism.

All opportunities offered by all objects in the world (including characters and environments) are published to a global list of opportunities. The character will choose one of these opportunities as its goal. To reduce the list to a single opportunity, a set of filters is used.
Each filter

• orders the opportunities according to a certain value
• reduces the size of the list to a fixed number starting from the top

The filters are executed in a specific sequence.
Every ordering includes some randomness that allows for a small chance that inappropriate items are included in the top of the list.

Happiness has been replaced by Comfort. The difference is that while Happiness made the character choose the opportunity that promises the most happiness, Comfort makes the character choose the opportunity that fits the relationship best.

Situation is a new parameter that is used for making sure that the character is in the capacity to execute an opportunity.

1. Filter of Situation: possible opportunities only

• removes the opportunities that the character cannot choose, due to its situation
• reduces the list to possible opportunities only

2. Filter of Distance: near objects first

• orders the remaining opportunities according to distance modified by object type (characters can be further away than objects)
• reduces the list to a maximum of 30 items

3. Filter of Relationship: loved objects first

• orders the remaining opportunities according to relationship modified by attention span
• reduces the list to a maximum of 15 items

4. Filter of Comfort: appropriate opportunities first

• orders the remaining opportunities according to match of relationship & promise modified by fascination
• reduces the list to a single goal

Thanks to Comfort, relationships can evolve more dynamically as a result of achieving a goal. Instead of simply improving, the difference between promise and relationship can be added to relationship. If the promise was larger than the relationship, it improves. If it was smaller, it decreases.

5 Comments »

Posted by Michael on June 9th, 2006, in Development

It’s funny how thinking about models to drive autonomous characters seems like a philosophical excercise sometimes. Maybe this is only because of the words we use. But we shouldn’t let philosophical preferences influence decisions about what might be effective artistic tools, should we? Consumerism, for example, might be a good model for Drama Princess. But it should be treated only as a starting point, not as a metaphor.

Likewise, we have to use the words we choose (happiness, relationship, fascination, shock, etc) for their symbolic meaning in the context of the design and not for their meaning in natural language. On the other hand, choosing appropriate words does help thinking about the design.

Anyway, there’s a bunch of things that I don’t like about the model described below. So this is definitely not the final design. It has already moved away from its original concept and I thought I’d publish it before it gets mutilated beyond recognition.

INTRODUCTION

A Drama Princess character moves through a world of opportunities. Every object and character in the world advertises the things you can do with it to a global list of opportunities. This list is ordered differently for each character, according to parameters defined by its personality and condition. The character doesn’t make a choice: it just does the thing that is at the top of the list. Doing this thing influences the character’s condition. This leads to changes in the order of the opportunities list as a result of which the character’s behaviour will show variation and growth. At each level of the decision making process, a certain degree of randomness is mixed in, allowing for unexpected behaviour.

Environment, decision making, action and consequences…

12 Comments »