i777777o61776eo636f6dz.oszar.com Open in urlscan Pro
172.67.220.76  Public Scan

Form analysis
2 forms found in the DOM

GET /search

<form action="/search" method="get" id="search-block-form" accept-charset="UTF-8">
  <div>
    <div class="container-inline">
      <h2 class="element-invisible">Search form</h2>
      <div class="form-item form-type-textfield form-item-search-block-form">
        <label class="element-invisible" for="edit-s">Search </label><input title="Enter the terms you wish to search for." type="text" id="edit-s" name="keys" value="" size="15" maxlength="128" class="form-text">
      </div>
      <div class="form-actions form-wrapper">
        <input type="submit" value="Search" class="form-submit">
      </div>
    </div>
  </div>
</form>

POST

<form action="" method="POST" style="all:unset;float:left;">
  <input type="hidden" name="URL_Adresi" value="https://www.awn.com/animationworld/shutterstock-and-nvidias-text-3d-modeler-helping-hand-sleep-deprived-artists">
  <select name="DNS_Adresi" onchange="this.form.submit()" style="all:unset;border:1px solid silver;border-width:0;margin:0px 0px 0px 20px;background-color:transparent;color:black;line-height:20px;text-align:left;padding:0 5px;font-size:14px;">
    <option value="0" selected="true">Otomatik - 192.249.121.68</option>
    <option value="11">CloudFlare DNS</option>
    <option value="12">Türk Telekom DNS</option>
    <option value="13">Google DNS</option>
    <option value="14">Open DNS</option>
  </select>
</form>

Text Content

Skip to main content



SEARCH FORM

Search


Visit AWN on Twitter Visit AWN on Facebook Visit AWN on YouTube Visit AWN on
Instagram Visit AWN on LinkedIn Visit AWN on FlipBoard Subscribe to AWN


Log In | Register
 * Home
 * News
 * Blogs
 * AnimationWorld
 * VFXWorld
 * Events
 * Forums
 * Jobs
 * More
   * Interviews
   * Reviews
   * Galleries
   * All Categories
   * AWNtv
     * Most Recent Videos
     * Top Playlists
     * All Video Categories
   * Animation Industry Database
   * Filmporium Store
   * How To Succeed in Animation


SHUTTERSTOCK AND NVIDIA’S TEXT-TO-3D MODELER: A HELPING HAND FOR SLEEP-DEPRIVED
ARTISTS

Releasing to the public in early fall, the asset platform company’s new gen
AI-based service, drawing from its library of over 500,000 ethically-source 3D
models and 650 million images, will provide creators the ability to generate
elaborate new models in seconds, helping them simplify their early design
process, bypass busywork, and focus on customizing and animating their
characters.

By Victoria Davis | Tuesday, August 20, 2024 at 12:43pm
In 3D, AI, CG, People, Technology | ANIMATIONWorld, VFXWorld | Geographic
Region: All

There’s plenty of caution and mistrust around the idea of artificial
intelligence intruding on animation production. There’s fear it will steal or
otherwise make redundant many animation production jobs, putting hundreds, if
not thousands of artists out of work. 

But Dade Orgeron, VP of 3D Innovation at Shutterstock, believes AI can work for
animators, not against them, and even make their lives easier. 

“I did 3D production for 15 years and, let’s face it, that involved a lot of
late nights,” says Orgeron. “Ideally, generative AI should at the very least
help you to get home on time. Let’s get through those design decisions and focus
on the hero content. Let's get you doing the things that you really like to do
and absorb the tedious tasks on the generative AI side. That's been our
philosophy from the get-go.”

Back in March, Shutterstock – a creative platform for high-quality assets, tools
and services – announced its partnership with NVIDIA, famous for, among a myriad
of things, creating applications for accelerated computing, in order to create a
text-to-3D modeler. Soon to be made available from Shutterstock for commercial
use, NVIDIA Edify-3D models will be trained with images and video from
Shutterstock and 3D assets from its subsidiary TurboSquid (acquired in 2021),
the world’s largest marketplace for 3D content. Once models are trained with
Shutterstock’s data, its service will use NVIDIA Picasso, a foundry for custom
generative AI models. Users will be able to use the Shutterstock service to
convert text prompts into high-fidelity 3D assets based initially on the
Universal Scene Description (USD) file framework.

Though the technology has only been made available to select clientele, the
general public will be granted access early this fall. In the meantime, anyone
can test out the modeler for themselves here:
https://i6275696c64o6e7669646961o636f6dz.oszar.com/shutterstock/edify-3d 







Currently, Shutterstock’s Generative 3D service is being used by partners like
NVIDIA along with HP, Trigger XR and the Creative Apprenticeship participants at
WPP, a British multinational communications, advertising, public relations and
technology company. 

“Being able to bring something from just an idea in your head to actually seeing
it on screen or on paper within a minute is incredible,” says Omotara Edu, a
creative technologist at WPP. “And I think that leaves you a lot of space to
then be more creative.” 

Shutterstock is also utilizing Discord’s services for beta testing. 

“We had some pretty good relationships already in place with NVIDIA going into
this, which I think made this a little bit easier,” notes Orgeron. “But one of
the things that really helped solidify the possibility of us moving in this
direction with NVIDIA was our approach to generative AI, which is more of an
ethical or responsible approach. That was one of the key things NVIDIA felt was
critical for them, and it was just in perfect alignment with where we were going
with our pillars of trust for AI.”

This not only included making sure all content the modeler drew from was
licensed data from Shutterstock, with nothing scraped from the web, but also
ensuring the generative modeler would hyper-enable creatives, rather than
disable their workflows or require studios and animators to change a single
thing about their pipeline. The modeler won’t cost jobs. It only simplifies the
preliminary design process so animators can breeze past the busywork and get
focused on the details of customizing and animating their characters. 

“We wanted to do that tedious, initial homework for you to really get this tool
to a point where it was truly useful and could accelerate workflows,” explains
Orgeron. “You work with the prompt, and you can generate a 3D model in seconds.
Normally, it takes hours to create a 3D model. And that’s not including the
painting and texturing process.” 







Orgeron, previously the chief creative officer at TurboSquid, says he and his
fellow company employees had always been proponents of standardizing the 3D
industry. The company was constantly looking for ways to make sure there were
interchangeable 3D formats that worked well across multiple creation platforms
and applications. But, he says, it’s always been an uphill battle. 

“But I think we’re finally at this point where this modeler is going to open the
door to truly democratize 3D,” shares Orgeron. “It will mean that people won’t
need to be experts at every single one of the complicated facets of 3D to work
and make a career. Before, if you had a model, maybe you didn’t know how to
texture. Or you could texture but not animate. Or you can animate, texture and
model, but there just aren’t enough hours in the day and you’re on deadline and
it’s either go home for dinner or work late and meet the deadline. Generative AI
can help with those different facets of animation.”

This AI too – with content storage that includes more than 500,000
ethically-sourced 3D models, more than 650 million images and detailed
Shutterstock metadata – isn’t just meant for producing character models. It also
includes generating HDRi background maps, a 360-degree image that includes
lighting information as well as visuals.

“As with anything generative, I wouldn't say it's at the final development
stage, even now,” says Orgeron of the 3D modeling tool. “But we’ve got PBR
materials – base color, a metalness map and a roughness – which are basic, but
it does give you some control over different material types. We also made sure
the mesh wouldn’t be triangle soup. We didn't want it to be something that
artists couldn't take to the next level if they wanted to. Because it’s in
quads, a good 3D artist can go in there, add more detail, adjust the textures,
all that. It’s a great starting point.”







He adds, “We saw this modeler as having two main uses. Maybe you just had some
ideas in your head, maybe you had an illustration already, or you had an image
already of what you had in mind, and you really wanted to bring that into
fruition as a 3D model. So, it's great for that. But then maybe also you just
wanted something to kind of fill up a scene. You've got this really complex
thing that you're making. You've got some hero objects that maybe you created
from scratch, or maybe you bought them at TurboSquid. You just want all the
other accouterments of that scene. You want to be able to fill that scene up.
That's where this can help you with as well. We’ve seen the quality of this tool
improve month after month and we’re nowhere near the end.”

Orgeron likens prompts to design work. The more specific language one uses, the
better the regenerative process will be. And this is not something just anyone
can do off the cuff. Creating a usable prompt that will generate something of
quality does require a well-honed knowledge of animation and what it means to
design a character or a setting. But Shutterstock and NVIDIA thought of that,
too.

“We do have a function on the prompt within the tool that will help enrich the
prompt for you,” Orgeron notes. “This can be turned off, but even if you choose
to keep it on, the details within the automatic enrichment can be changed and
modified so it’s more in line with what the animator is going after. The prompt
can continuously be refined.”

After solidifying the first draft of a prompt, a model will generate in roughly
10 seconds. Even after the model has been generated, the prompt can continue to
be modified and adjusted. The customization capabilities are limited at the
moment but, eventually, Orgeron says they’d like to break those barriers down as
well. 

“Eventually, since this is a foundational model, we’d like to make a way for
creatives to add their own content into the modeler and be able to fine-tune it
so it fits the exact style they’re looking for and so the modeler can continue
to generate in that style,” he explains.







Orgeron continues, “In its simplest form, you could use this now to just
generate an asset and download that asset as an OBJ, a POI, a GLB or a USD. So,
if you want to build a library of assets that your artist in your studio could
all pull from, you could build up that whole library using the generator and
store that wherever you normally would store your projects and then utilize that
content. Further down the line, if you did want to fine-tune or train your own
model, so that there was only content in that specific style, that would become
yours. It would be siloed off to you. That would be your model, based on our
foundation model, that only your team would have access to, so no one has to
worry about an IP being stolen from them. So, then you end up with not only a 3D
modeler, but a content processing generator.”

Having showcased the technology at SIGGRAPH this year and NVIDIA’s global
artificial intelligence conference GTC, Orgeron says seeing the positive
reactions from people in the industry, getting feedback that the capabilities of
the modeler were “unlocking new ideas” in filmmakers heads, was an incredible
reward after TurboSquid spent the last two decades trying to get a product like
this out to the masses. 

“It took a while and was a lot of hard work, but I’m really thankful for our
partnership with NVIDIA because that’s the kind of firepower you need for
pulling something like this off,” says Orgeron. “It’s not easy.”

Victoria Davis is a full-time, freelance journalist and part-time Otaku with an
affinity for all things anime. She's reported on numerous stories from activist
news to entertainment. Find more about her work at victoriadavisdepiction.com.

Tags 
New GenAI-Based Creative Tools
Shutterstock
nvidia
TurboSquid
GenAi
Generative AI
AI
Artificial Intelligence
3D models
3D animation
CG
cg animation
Dade Orgeron
NVIDIA Edify
NVIDIA Picasso
text-to-3D
Technology
HP
Trigger XR
WPP
Omotara Edu
 * 
 * 
 * 
 * 
 * 
 * 
 * 

More From ANIMATIONWorld:
Next Post
The Haunting Animated Prologue in Fully Restored ‘Caligula: The Ultimate Cut’
Previous Post
‘The Filipino Story’ Explores the Philippines’ History, Culture and Social
Dynamics
Related Content 
Shutterstock Brings Generative AI to 3D Scene Backgrounds with NVIDIA Picasso
Shutterstock Teams With NVIDIA on Generative AI 3D Artist Tools
Shutterstock Announces NeRF AI Tech Collaborations
NVIDIA Launches NVIDIA AI Foundry
New White Paper Dives Deep into NVIDIA Omniverse Enterprise in Animation and VFX



 


LATEST AWN MUST READS

How Japan’s ‘Age of Robotics’ Inspired ‘Terminator Zero’
Ingenuity Studios Corrals the VFX in ‘Horizon: An American Saga Chapter 1’
Binocular Briefs - August 2024
From Struggle to Strength: How ‘47’ Was Made For and With the Down Syndrome
Community
The Haunting Animated Prologue in Fully Restored ‘Caligula: The Ultimate Cut’
DNEG’s Fury in the Wasteland: Key VFX on ‘Furiosa: A Mad Max Saga’
See More »


TOP HEADLINE NEWS

HAELE 3D - Portrait Studio Pro Coming to Steam
3Dconnexion Shares ‘SpaceMouse Wireless – Bluetooth Edition’ Case Study
Proof Shares ‘Damsel’ Previs, Postvis Reels
V-Nova Launches V-Nova Studios Division
Prime Video Releases ‘The Legend of Vox Machina’ Season 3 Trailer, Key Art
The Salvation Poem Project Unveils Animated Feature Film ‘Light of the World’
Ánima Kitchent’s ‘Cuquin’ Season 2 Coming to Max and Cartoonito Latin America
‘Hanna-Barbera, the Recorded History: From Modern Stone Age to Meddling Kids’
Now Available
See More »



CAREER CONNECTIONS

 * Two Tenure/Tenure-Track Faculty Positions in Applied Virtual Imagery and
   Extended Reality for Interdisciplinary Innovations - Indiana University
   Bloomington
 * Creative Director University of Georgia
 * Production Artist Medpace
 * Art Director, (Sports & Entertainment Motion Design) - CW Network - Burbank,
   CA Nexstar Media Group
 * Associate Director, Graphic Design & Animation Patient Engagement, CDDA Lilly
   and Company
 * Senior Motion Designer Nexstar Media Group
 * VCUarts Adjunct Instructor - Department of Graphic Design Virginia
   Commonwealth University
 * Graphic Designer The University of North Carolina at Charlotte

See More »


RECENT COMMENTS

 * Shakeem winn
   
   Dude lived a long and fruitful life. R I.P
   
   Japanese Visual Artist and Animator Keiichi Tanaami Passes at 88 · 2 hours
   ago

 * StevenPettway93
   
   At MIPTV, the incredible film showcased its ambitious animation slate,
   highlighting groundbreaking projects set to captivate audiences worldwide.
   Among the notable highlights was the announcement...
   
   Incredible Film Reveals Animation Slate for MIPTV  · 5 days ago

 * Emilio Romero
   
   I don't know if it's just me, but to hear Helm say, "My pride, my joy. You
   could rule the world" isn't the kind of things heroes are supposed to say in
   Middle Earth. But...
   
   Long-Awaited ‘The Lord of the Rings: The War of the Rohirrim’ Trailer
   Drops · 5 days ago

 * Bible Guy
   
   Is he going to include the rape scene from the comics or not? That would be a
   great question to ask, me personally I think he should because it will help
   bring light to men who were raped by women!...
   
   A Talk with the ‘Invincible’ Robert Kirkman · 1 week ago

 * Justin Grogg
   
   Will they cast the same people and will clover make an appearance in this
   series
   
   Disney Jr. Greenlights ‘Sofia the First: Royal Magic’ · 1 week ago

 * Osman Hossain
   
   I love Sofia The First reboots on Disney Junior, Disney Channel and Disney+.
   
   Disney Jr. Greenlights ‘Sofia the First: Royal Magic’ · 1 week ago




MOST POPULAR

 * Long-Awaited ‘The Lord of the Rings: The War of the Rohirrim’ Trailer Drops
   
   1 comment · 5 days ago

 * Disney Jr. Greenlights ‘Sofia the First: Royal Magic’
   
   2 comments · 1 week ago

 * Brad Bird Developing ‘Incredibles 3’ at Pixar
   
   1 comment · 1 week ago

 * Cloud Computing Is Stifling Creativity in VFX Studios - Here’s How to Fix It
   
   1 comment · 2 weeks ago

 * Wonderstorm Launches ‘The Dragon Prince: Xadia’ RPG
   
   1 comment · 3 weeks ago


ELSEWHERE ON AWN

 * FOX Renews ‘Krapopolis’ for Fourth Season
 * HBO Renews ‘House of The Dragon’ for Season 3
 * ‘Sausage Party: Foodtopia’ Renewed for Second Season at Prime Video
 * Prime Video’s ‘Invincible’ Renewed for Fourth Season
 * ‘Camp Snoopy’ Renewed for Second Season at Apple TV+
 * FOX Renews ‘Universal Basic Guys’ Prior to Season 1 Debut
 * Peacock Renews Seth MacFarlane’s ‘Ted’
 * MTV Greenlights ‘Mike Judge’s Beavis and Butt-Head’ Season 3


ANIMATION WORLD NETWORK

Contact Us | About Us | Terms Of Service | Privacy Policy | Media Kit

© 2023 AWN, Inc. AWN.com - Part of the Animation World Network - 13300 Victory
Blvd. Suite 365 Van Nuys, CA 91401




Otomatik - 192.249.121.68 CloudFlare DNS Türk Telekom DNS Google DNS Open DNS
OSZAR »