Under The Skin (2014)

Description

Of all the films I have worked on recently, I have a feeling this one will still be being talked about in decades time. Hugely divisive when it premiered (audiences either cheered or booed when it was was shown at Venice Film Festival), this movie has very little dialogue or obvious plot arcs and certainly no spoon-feeding exposition to the audience.
I worked at One Of Us several times over a couple of years whilst this made it’s long journey through post production. Director Jonathan Glazer was insistent that the films vfx sequences looked as little like vfx as possible – a good deal of the movie was shot with hidden cameras and non-professional actors, and so this raw feel would have to blend into the abstract, science-fiction scenes without feeling jarring.
My shots were all in the final ‘reveal’ sequence, when Scarlett Johansson’s seductress character sheds her outer human skin to reveal the alien beneath. This involved augmenting an incredible prosthetic outfit with fully CG body parts. The challenges came from the way in which human skin and prosthetic rubber stretch in incredibly nuanced ways – then having to track this stretching and build comp tools that stretch the CG renders to match. The shots in this sequence were also framed and edited in a very unforgiving way – lots of long, locked-off close-ups that give you nowhere to hide any mistakes.
In this interview for Film4, Tom Debenham and Dom Parker (directors at One Of Us and VFX supervisors on Under The Skin) talk in more depth about the challenges of the end sequence

D: The biggest challenge for us was the end sequence where Laura peels off her human skin and reveals herself underneath; the black agent in disguise. It is very much a hybrid of prosthetics and performers. There’s Scarlett Johansson as you see her in the film, but there’s also the skinny alien.

The head is entirely digital at the end and it’s based on what Scarlett’s static face is like – a death mask really – and that surface is entirely digital. And then the skin underneath is treated to match the balance between those surfaces. There is a lot of very, very precise animation, not the sort of thing that you would say ‘oh, that looks like a computer generated character’, but tiny things about the way the head was positioned and the nuances of the movement on the neck were all incredibly precisely calibrated over a very long period of time. It’s not normal animation, it’s very miniscule adjustments, to get the tone right, because it had to be very still and very focussed and should absolutely not trip the audience out of it or have them thinking for even a moment that anything other than what we were seeing portrayed was happening.

Tags: Compositing