Friday, August 31, 2007

Possible Research Areas

Here's a couple of ideas that we could look into:

1. The psychology of gesture - in animation, we work a lot from reference footage and learn how exaggerate certain gestures, but there's not much time devoted to what certain movements and gestures mean on a psychological level i.e. do gestures made with the palm facing up have a certain embedded meaning? If the answer is yes, then this information could potentially be useful to animators.

2. Performance capture - when it works, when it doesn't. Why are keyframe animators able to express emotion so succinctly when we are still struggling to duplicate the same using capture technologies? Does the answer lie in more technology, more cameras and more sensors, or are we missing something entirely? Scott McCloud's work on comics as a graphic art might potentially be relevant here.

3. Gesture and culture - are there gestures that transcend cultures - sort of a universal language? There have been studies that demonstrate that humans do exhibit certain innate tendencies towards language - could the same principle possibly apply to gesture as well?

Any other ideas about what could be interesting for us to research?


Ryan said...

A few ideas:

1. The illusion of intelligence in non-playable characters (NPCs). Perhaps there could be an investigation into the chronology of videogame A.I development in regards to the increase in the number and variety of response types (both in body language, and verbal responses).

2. Idle states. What does a characters do when they do nothing and they're just waiting (or listening)?

3. Risking life and death. Compare the emotional response of both CG and live-action characters in situations of mortal danger. Do audience members show a decrease in emotional investment if the character is CG? How can animators put the fear / risk / anxiety back into animation? -(perhaps we could create a series of CG motion studies of characters in danger and let vistors rank their emotional response).

Arjun said...


I think these are really interesting. The last one can tie-in to the idea of the uncanny valley. I like the video game idea too - it's a different approach to helping us understand what makes an action believable.

Ryan said...

another idea:

1. Humanoid characters vs non-humanoid characters. Since the uncanny valley only applies to synthetic humans (I think), we could investigate the difference in psychological acceptance and emotional investment in a student test group by creating (again) a series of motion studies for site visitors to rank.

Hyun said...

1. What the brain ordering to the body:
Even in a simple movement, depending on the brain's decision, our body parts move in different order. For example, when you hear someone calling your name in your back, your face turns to the back the first- to check where the sound from, then the upper body turns the next- to help to get better vision, and very slightly the pelvis follows the turning- this could be follow through. But if you were playing a game in the same situation, you could move your body first- it can be a immediate body reaction, and then turn your head the last- if you were really into the game. To me, one of the most difficulties on animating characters was how to set the order of each body movement. Depending on the order of movements, you could make different characters. It could be interesting and helpful to make a list of body parts action order and the possible characters from the order.
2. Multiple characters dancing:
On most social dances, a man and a woman make a pair and dance together. Even though they don't make any plan before dancing, many of them can dance in perfect harmony. Mostly the leader gives a sign to the follower before they make a move. It could be interesting to research what kind o f signs make what kinds of reactions.

Hyun said...

wow, I totally feel like animation is very scientific study area.

Ryan said...

Right, so I'm doing something regarding games, interactivity, and automated character animation (gestures, performance, the works).

I've researched and tested a way to develop and post games that can be played within Internet Explorer. Very exciting. Theory into practice baby. And yes, it's Blender.

I'm going to start compiling my research and tests on a webpage that I'll create shortly (for experimentation, review, and critique).

So basically, I'll do some research, find some interesting questions, and make an interactive experience that addresses those know...furthering the research. That's the idea right? I mean, this isn't a book report.


Arjun said...


Sounds good. Hyun and I are looking more into how gestures are classified and deconstructed into pathways, with the idea that we could come up with a simple visual representation that animators can reference. I'm also going to look at whether there's anything on the neurological front that helps us understand how gestures are wired to the brain.

Also, I'll create a separate thread for you to post your findings about games.

chiachi said...

About the research,I do have insteresting in the both.I don't have much experience about the movement in animation.I just start to learn.So I don't know which one is better to me.

Ryan's research:It's interesting,but I don't have much experience with the game character.The idle state
sound funny to me.but I don't have idea of investment now.Can I have more time to think about?
Hyun&Arjun's reasearch:It's also interesting.Does the reflex action have associate with "brain order to the
or "what kind of signs make what kind of reaction." I think I like this research.Also,how to get a balance between the
camera and kinetics.what vision is best for audience to feel?But I don't have any idea about this now.

chiachi said...

I have a idea.what about the center of gravity when we do some action.For example,when we are walking,we move the foot ,hand and shoulder and butt.Which one is the center of the movement.Is the center one faster then others?will different center show different movement and emotion? How is the relation?Is it like leverage?I don't know if the research is too wide?or too simple?

Also,I think I am interested in "What the brain ordering to the body". maybe I can join the research?

Arjun said...


About the center of gravity - I feel that Richard Williams, Preston Blair, etc go into this stuff in some detail so we have to think a little bit more carefully about how we differentiate it from their work. Off the top of my head, I'm not sure of a good angle, but perhaps someone else in the group can up with up.

As for connecting the brain and gesture, I'm very interested in that area myself, so we can meet and come up with some goals for the next few weeks. I'll create a separate thread for it.

Tim said...

Hey All,

Wanted to post a few ideas i have been thinking about. There may be parallels between many of the items that have already been discussed, but please comment and let me know your thoughts so i can limit my research to incorporate a certain "area" of animation:

- What brings a character to life? what is that IT Factor? Its one thing to make a character move, but what establishes the natural "living" illusion of the character (exaggeration, personality traits, body weights, relatable motions, etc) What pushes animations beyond moving drawings and surfaces on the illusion of being perceived as REAL characters?

- To what extent can character animation be stretched in terms of believability, in regards of motion, squash & stretch, contortions of anatomy. Is there a limit? or can anything be done as long as the chracter "feels" real and "natural" What is the key to establishing this Natural Quality?

- When things are animated as close to realism as possible or are developed through motion capture to relay exact motions of anatomy, why is there a lack lust quality? Something usually reads as feeling "off", not quite alive but rather artificial. May stem back to this "lifelike" natural quality....what is it, what is required to achieve it?, what is being done?

- To what extent does it take for an animated character to evoke an emotional response from an audience (crying, fear, lust, annoyance ,etc.) How can an animator break the idea that the character is just imaginary or fictional and evoke a relatable "real person" sense to the audience. Can this be done without story? Or without facial gesture?

-Performance: what makes it successful. Why are great performances/actors so great. What qualities do they possess that allow them to transcend celebrity and "become" this character they are portraying. How are animated characters mimic-ing or exagerating acting techniques in order to perform successfully. Ways to mold a characters personality soley based on movement and gesture design.

-Perhaps a study on how EXAGGERATION plays a major role in establishing certain movement in animation. How this kind of movement establishes animated characters to thrive on another plane of motion existance not able to be mimiced in live action. Comparisons of how normal actions are percieved next to actions which are exagerated and how they are percieved?

Well, there's the summary of my thoughts for the last week. Perhaps some structure will be formed of these ideas. I would love some feedback in regards to some of the topics. As mentioned earlier i am still trying to find a solid area to study within the realm of character animation since much has already been technically explained by great animators of the past. I don't want to research something that has already been explained elsewhere but these are just thoughts that i found provoking and interesting to learn about. Thanks again guys! :)

PS: Sorry for length :)

Arjun said...


I think you have some interesting ideas. I think the questions you have asked are fairly broad, so in order to narrow them down, here are a couple of thoughts:
1. Try to define what is a character and what is character animation. There will probably be a few different definitions here, but it's worthwhile to do this at the outset.

2. Look at Giotto and the Arena Chapel
According to Kathy, his work represents a pivotal shift in the portrayal of gesture in art, and will help us understand the IT factor that you refer to.

3. Similarly, Scott McCloud's work in 'Understanding Comics' talks about abstraction in comic art. I feel that these concepts can be applied to gesture as well. You can borrow my copy of the book.

Does that help?