Technology puts back Hollywood’s retirement age

Last November Netflix successfully premiered the movie “The Irishman”, produced and directed by Martin Scorsese and based on Charles Brandt’s biography of Frank Sheeran called “I heard you paint houses”. A star-studded cast tells Sheeran’s story. Sheeran, played by Robert De Niro, was a trucker who became a hitman for the gangster Russell Bufalino, played by Joe Pesci, and the union leader Jimmy Hoffa, played by Al Pacino.

Quite apart from this stellar cast, Scorsese chose the company Industrial Light & Magic (ILM) for production of visual effects (VFX) and computer generated imagery (CGI). Founded in 1975 by George Lucas, the company has taken on visual effects in some of Hollywood’s best-known super productions: Star Wars, Titanic, Jurassic Park, Harry Potter, Pirates of the Caribbean, Avengers and many other blockbusters. This company is one of the US’s biggest VFX producers, turning over 450 million dollars a year. Even this striking figure pales into insignificance against the valuation of 259 billion dollars placed on the animation, visual effects and video-game industry in 2018. It is also an industry that has been growing exponentially in recent years thanks to more affordable internet access, the mass takeup of smartphones and video-on-demand systems.

The truth is that the visual effects industry, maybe because it sits as a perfect, creativity-kindling intersection between art and technology, is a constant source of innovation. Prime examples might be the coloring of video footage to give spring-like hues to scenes actually shot in January, fluid simulation to render water or explosions, artificial intelligence to animate a pod of killer whales in which each one has its own behavior or movement-capturing technology to animate a virtual avatar, to mention only some of many.

In the case of the “Irishman”, ILM developed a groundbreaking system to make the actors look younger. For those of you who haven’t seen the movie, it takes place from 1949 to 2000 and is set at three different moments of the characters’ lives. At his youngest, for example, Pacino is 39-years old. Until ILM took on this challenge, there were two main technologies for making actors look younger.

The first involves creating a doppelganger using images of the actor’s face making different gestures. The Mova camera system is an example of systems of this type. Once the actor’s double has been generated it is superimposed over the actor’s image and the production team then animates the double manually, modifying its characteristics. The downside of this system is that the acting tends to lose realism and authenticity, since it is now the technicians who manually animate the virtual double rather than the actors actually playing the part themselves. The movie “The Curious Case of Benjamin Button” for example, starring Brad Pitt, used this method. The whole process, involving a team of over 155 people, took two years to compete.

Bejamin Button

The second technology, like the first, starts with the creation of a virtual double of the actor, by scanning the face and then modifying it on the basis of performance capture. In this case, however, the virtual double was animated by the actor himself on the basis of his facial expressions. For this purpose the actor was fitted with a movement-capture device comprising a helmet with several cameras focused on the face and with tracking dots all over the face. The camera device picks up the movement of these tracking dots on the face, animating the digital double accordingly. The downside of this system is that certain facial expressions are lost (tracking-dot precision problem), so the acting suffers in quality. It also obliges the actor to wear a helmet with several cameras during the shoot and with his or her face plastered with tracking dots, hindering the performance further. This is the system used by the recently released “Gemini Man” with Will Smith.

Will Smith

In “The Irishman”  Scorsese wanted his actors to be able to perform freely without having to wear cumbersome equipment or work with their faces covered in tracking dots, but without forfeiting a single detail of the actor’s performance. ILM therefore set out to develop a new system comprising 1) a set of cameras affectionately dubbed the “three headed monster” during the shoot, and 2) image processing software called Flux.

THREE-HEADED-MONSTER-1

The camera rig has different components. At the center is a RED Helium 8k digital camera to record the video footage. Each side of it are two mini high-resolution ARRI Alexa cameras, called witness cameras, modified to register only infrared light and synchronized with the central camera. This system captures the actor’s facial performance from two angles, generating an infrared stereo image. The scene is also lit with infrared spotlights to prevent shadows in this spectrum; this light is invisible for the central camera. This lighting helps the witness cameras to capture the actor’s face volumetric information better without using tracking dots.

As well as these three cameras there is also a previous preparation process involving a LIDAR scanner to obtain the exact position of all the points of natural and artificial light, while the whole scene is captured in HDRI to obtain the light intensity and color of these points.

This whole data set is obtained for each frame of the sequence being shot and is then sent to the Flux software for processing. This software combines the infrared images with the real image, the scene lighting and shading, to create a virtual mesh superimposed over the actor’s real face. This mesh is a geometric model of the actor’s face for each image and is used to deform the actor’s young digital double.

Máscara

This digital double is created previously by two commercial teams 1) Disney’s Medusa for capturing hundreds of poses of the actor’s face, and 2) Light Stage software for capturing the skin, texture and porosity details. From the contemporary digital double, four models are then created for each actor at different ages, basically by painting textures, adding or removing wrinkles, etc.

Robert de Niro

Although the Flux software is a brilliant specimen of data fusion, there is also a fair amount of post-production manual work to correct any discrepancies in the actors’ young doubles. ILM created two new tools to help out in this manual post-production process.

First of all a catalogue or repository of images is built up from each actor’s past movie career with details of nose, mouth, eyes, etc. This catalogue provides a detailed reference bank that can then be used in post-production for phasing in “young” facial elements in the sequences where this is needed.

Secondly, a machine-learning tool was created to scan the repository and find the video footage or image most resembling the sequence currently being worked on, according to varied criteria such as: lighting, angle, eyes, etc. Swift access to these images helped the post-production team to correct any discrepancies generated by Flux.

Robert de Niro

One of the main advantages of ILM’s new system is that it is non-intrusive and can be used for close-up shots where performance capture is important, unlike other deaging techniques used hitherto, where the emphasis is on action. It also enables the scene to be shot with the actors naturally together on the set instead of in a special movement-capturing set or wearing special cameras with the face spotted with tracking dots.

Maybe the movie wasn’t to your liking, and we might all have our own views about Netflix’s shenanigans to be able to take part in the Oscars, but ILM has without doubt dreamed up a new technique that is now here to stay. Who knows if further miniaturization and automation might be possible in the future and incorporated into out smart phones? It certainly looks likely to be the most widely-used filter in Instagram.

 

Author: Víctor Gaspar Martín

 

Las opiniones vertidas por el autor son enteramente suyas y no siempre representan la opinión de GMV
The author’s views are entirely his own and may not reflect the views of GMV
Share

    Leave a Reply

    Your email address will not be published. Required fields are marked *

      I accept the privacy policy Acepto la Política de Privacidad

    Basic Data-Protection information:

    Data-protection supervisor: GMV Innovating Solutions SL
    Purpose: Answer questions, respond to user complaints and recommendations, receive job applicants resumes and career information.
    Legitimation: Consent of data subject
    Addresses: Grupo GMV companies
    Rights: Access, rectify and cancel data plus other rights, as explained in additional information
    Additional information: You can check out the additional and detailed data-protection information on our website:Privacy Policy

    You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

    WP-SpamFree by Pole Position Marketing

    3,516 Total Views