HBO Westworld Main Titles

HBO’s Westworld main title sequence, created by Elastic, opens the network’s high-concept reimagining of Michael Crichton’s 1973 science fiction thriller. Directed by Patrick Clair, the titles show robotic tools stringing together a galloping horse of ribcage and sinew, a half-formed pale rider taking the saddle, two unfinished lovers embracing, and a player piano performing without its player — imagery evoking Leonardo da Vinci, Mamoru Oshii, and Chris Cunningham while hinting at the ethical and existential themes of the series. From producers Jonathan Nolan, Lisa Joy, and J.J. Abrams, the show spans both the ersatz Wild West of the park’s cybernetic “hosts” and the glossy near-future of its custodians and creators.

SCANable performed LiDAR and 3D scans of sets, actors, props, and other items for this beautiful main title sequence, providing the scan data that helped Elastic craft its haunting digital imagery.

Taking a body apart requires having the whole body first

The sequence shows anatomy mid-assembly — a ribcage before muscle, a rider only partly formed, hands that stop at the wrist. Imagery like that depends on there being a complete and correct figure underneath to subtract from.

A partial form sculpted straight from reference tends to be expressive and anatomically loose. Working from a measured human and removing material keeps proportion, mass, and joint placement honest. That accuracy is what makes the incompleteness read as manufacture rather than as a mistake.

A synthetic look is built on real measurement

The unsettling quality of the imagery comes from precision, not from distortion. Machine-made things are exact, and a viewer registers exactness even when the subject is fictional.

Photogrammetric capture of performers gives that exactness for free. Skin, bone landmarks, and the way a shoulder actually sits are recorded rather than estimated, and the stylisation is applied on top of a correct base. See cyberscanning for the capture side and digital doubles for what it produces.

Design pipelines use scan data differently than shot pipelines

A title sequence is developed like design work. It is iterative, art-directed, and still finding its shape late in the process, and the same asset may be lit, shaded, and reframed a dozen different ways before anything locks.

That puts a premium on assets that are clean and reusable rather than tuned to one plate. Sensible topology, predictable UVs, real-world scale, and consistent naming let a design studio explore without rebuilding. Retopology and render-ready modeling is where that discipline lives.

Sets and props measure differently than they photograph

Photographic reference records how a set looked from one position under one lighting setup. It does not record how large anything was.

LiDAR of sets, props, and standing pieces produces the other half — dimensioned geometry a CG artist can build against, place objects inside, and match a virtual camera to. Our set and location scanning covers that spatial record, and it stays useful long after the stage has been reset.

Browse more television VFX work, or see the services we offer.