Research Project: Artifact 3: Uncanny Valley

Why Am I doing it?
Gage audiences reactions and receptions of differing forms of animation and character appearances. The more stylized the more acceptable? Does hyper-realism and inconsistencies cause dislike and detachment.  

How am I doing it?
Controlled focused groups with a large variety of participants. Completion of video-clip experiment to gather and plot statistics

Results

 

When Uncanny valley is applied to special effects in movies, the implications are clear: if a filmmaker strives for high levels of realism in CGI characters, they risk taking the humanlike resemblance too far, causing viewers to notice mismatches in characters appearance and movement.

Our emotional response to photo-realistic characters is that of unease and repulsion, not pleasure or likeness. This can be seen in the plummet of the purple likeness curve to rise in the green realism curve on the graph, This concerns the near photo-realistic characters of Beowulf, polar express and Imagemetric's Emily

However If a filmmaker decides to create characters in a more stylized manner, clearly signaling that they are not supposed to appear “almost human,” (The Incredibles and Avatar) we are more likely to view these characters as likeable than the characters designed to look photo-realistic. We see evidence of this with Avatar and it's 8.5 peak of likeness just before the plummet into the Uncanny Valley

"The alien Na'vi were humanoid and extremely lifelike, but they were blue-skinned with other clearly non-human features, so they didn't trigger the uncanny valley effect"

My results seem to largely support the theories of Uncanny Valley. Adding more and more human features to artificially created characters (Hal - Avatar) seems to increase there likeness and familiarity, but only to a certain point (Emily), before becoming too perfect and plummeting into repulsion and the valley.

Interesting enough it appears that "The Adventures of Tintin" breaks the rules and climbs out of the uncanny valley, with an incredible high level of realism to likeness ratio. This could represent a huge advance in motion-capture capture technology where the Uncanny valley theory is no longer applicable. where audiences are no longer effected by the in perfections and trickery which has triggered the uncanny feelings of dislike in the past.

Research Project: Artifact 2

Why am I doing it? 
Knowledge of the industry and it alteration over time, How visual fx pipeline works. Skillset of the worker, future of special effects and how they they're going to intersect.  Identify key issues which may result in lessened quality...

How am I doing it?
Conducting interviews and gathering professional idustry opinions

Results:
The hand-on, 'getting dirty' aspect of physical effects is missing entirely, the direct formative activity of touch and feel is gone, yet as far as what their trying to accomplish it's not that different. Every project still starts with drawing classically and VFX practitioners with traditional art skills in anatomy, lighting and experience in creating shapes and surfaces are still core to creating quality work... it's still the best medium for beginning work. 

CGI used to be deemed quicker, easier or cheaper than actually building props, yet has retracted. With recent software enhancements CGI can now be considered more complex and be much more expensive than physically producing an effect. Realism is paramount to the success of FX's, every effort is taken to avoid audiences realising flaws, the more physical reference material to intersect and time spent on set connecting with lighting the better the outcome for photorealism. 

The main culprit for breaking suspension of disbelief is still currently CGI humans with their uncanny qualities, everything can be lost within facial expressions and the eyes in particular. Workers are still in control of their tool and believe very little of their work to be automated or restrictive, work which is automated is usually reserved for background visuals which don't require much attention. Incredible amounts of skill and attention to detail still goes into foreground work and is the individual creation of the worker. 

For the future, many artists do believe that models have a greater impact and add a nice aesthetic and character to a film, yet they do view it as a dying art. Once the final frontier of human expression are cracked, with the trend and pace of current technology, there is no reason that within 5 years CGI will not have not totally eclipsed physical effects. However cgi could become too costly and retract.

 Key points - 

• Direct involvement of touch and feel is missing but traditional art skills are just as relevant if not more than the digital software knowledge 

• Cgi was once seen as a 'quick fix', easier option, but now it is in-fact much more complex and expensive. 

• Photorealism is the focus of many animators, physical reference is helpful in this regards and shouldn't be replaced 

• CGI humans are the main culprit of failed suspension of disbelief. 

• Animators are 'working on the tool' not 'through the tool', it is a direct formative activity and the final product is a result of the individuals creativity and skill, not mechanical operation.

• CGI will encapsulate physical effects