Upland partners with AI studio Kaedim to accelerate metaverse asset production

Upland’s experimental collaboration with the artificial intelligence (AI) startup Kaedim has proven to be more fruitful, with the metaverse platform seeing a 70% increase in 3D asset production speeds. 

It’s an outcome that will enable Upland to populate its immersive virtual world more rapidly while accelerating its pace of innovation. 

The partnership represents another key evolution for Upland, which began as a fairly simple, blockchain-based virtual property trading game. With Upland, players can buy, sell, and trade virtual properties mapped to real-world locations while engaging in an economy that caters to entrepreneurs, content creators, and gamers alike. 

The concept has caught on, with Upland seeing more adoption after being announced as FIFA’s official metaverse partner during the Qatar World Cup in 2022. 

At the last count, it had over three million registered players who traded 4 million digital assets. 

AI-enhanced asset creation

Digital asset creation is one of the most vital elements of Upland, as the game revolves around user-generated content. Players can design, manufacture, and trade digital goods and experiences with Upland. But to enable this, Upland must provide players with a consistent and easy-to-use pipeline to design and build novel digital assets. 

Due to Upland’s growth, its creators decided players needed a superior way to accelerate the production of digital assets on its platform, and that’s where Kaedim comes into the picture. 

The AI startup has created a design platform that uses machine learning algorithms to accelerate the 3D asset design process by up to 5-10 times compared to manual processes. 

Its designs are incredibly detailed, with features like automatic texturing and levels of detail that enable the creation of lifelike animations. 

Kaedim’s platform is powered by generative AI. The same technology underpins sophisticated AI models such as ChatGPT and Stable Diffusion. By integrating its image-to-mesh technology, Upland streamlined its 3D asset production process, reducing the time it takes to design and build more complex objects by several days.

Accelerated production 

The asset creation process in Upland is almost entirely automated, with Kaedim’s AI responding to user inputs to create an initial sketch of each new asset, which can then be refined through further inputs. 

Once the user is satisfied with the initial design, they can bring it to life with Kaedim’s AI modeling capabilities, creating meshes and textures to enhance the realism of each asset. Users can then import those refined assets into the Blender platform to add any final touches they desire. 

Kaedim detailed some impressive numbers, saying that Upland could reduce the time it took to design and produce 70 Totem assets from 245 days to 70 days. 

Whereas one Totem model took an average of 3.5 days to produce, it now takes less than a day. All told, Upland cut the time spent on designing and building 5,000 Totem life-forms by 72%.

In a statement, Upland said:

“In less than two weeks, we successfully took two original IP campaigns from inception to production, covering Map Assets, Structure Ornaments, Go Karts & Wearables.”

Paving the way for innovation

The more straightforward asset creation pipeline means that Upland users will benefit from a more rapidly evolving experience, as creators can introduce frequent updates for digital assets. 

Users can quickly incorporate feedback and suggestions to improve their assets by facilitating a more responsive development process. 

With these more visually striking assets, Kaedim has taken customization to a new level, allowing Upland users to personalize their in-game spaces and avatars with an ever-expanding array of content. 

Moreover, the accelerated production means that Upland creators have more time to collaborate with their peers and explore new ideas and concepts through brainstorming, increasing the pace of innovation within the game.

Follow Us on Google News