Synthesis AI, a San Francisco-based synthetic data technology company, has announced the development of a new text-to-3D technology that allows users to create realistic 3D digital humans using text prompts.
The technology leverages generative artificial intelligence (AI) and visual effects pipelines to create high-resolution digital humans that could be suitable for use in gaming, virtual reality, film, and simulation applications.
Synthesis AI’s text-to-3D technology allows users to input descriptions of the desired digital human, such as age, gender, ethnicity, hairstyle, and clothing. The machine then generates a 3D model that matches the specifications.
Using sliders, users can further edit the 3D model, adjusting features such as facial expressions and lighting.
Synthesis AI claims to be the first company to demonstrate text-to-3D digital human synthesis at such a high level of quality and detail.
Synthetic data mimics real data but does not include any personal or sensitive information. Compared to traditional methods, the text-to-3D capability saves a lot of time and money and improves efficiency.
“The text-to-3D capability we’re showcasing in synthesis labs takes a programmatic, API-driven approach as its starting point, adds a dead-simple prompt-based user interface and outputs a high-resolution 3D model that can be used as synthetic data across a broad range of use cases that require digital humans,” Yashar Behzadi, CEO and founder of Synthesis AI told VentureBeat.
Synthesis AI Has a Library of 100,000 Digital Humans
Synthesis AI has a proprietary library of over 100,000 digital humans (IDs) that supports its model training.
The company has also developed in-house models leveraging diffusion-based generative AI architectures to generate a diverse array of meshes, a collection of vertices, faces, and edges that define an object, governed by gender, age, ethnicity, and more elements.
Texture layers are created using a separate generative model with fine-grained independent control. Finally, a comprehensive and high-resolution 3D model gets produced by merging these two essential components.
“Creating a diverse set of humans is further complicated by the logistics of recruiting specific individuals and obtaining waivers,” Behzadi said.
“Starting with an inexpensively synthesized digital human are orders of magnitude faster and cheaper than either of those options. The text-to-3D capability enables on-demand generation of high-quality assets, saving weeks of time and thousands of dollars per model.”
The text-to-3D digital human capabilities of Synthesis AI will be available for beta testing to a select group of users by Q2 2021.
By making this capability available to external users, Synthesis AI plans to leverage the community’s feedback to refine the underlying generative models.
Behzadi claimed that this innovation will accelerate the development and reduce the costs of 3D applications in various sectors such as VFX, automotive, smart cities, industrial, and manufacturing simulations, and AR/VR.
Image and Art AI Generators Hit the Market Amid AI Frenzy
The recent AI frenzy has seen a rush among different tech companies to unveil their AI-powered image and art generators.
OpenAI’s DALL-E, a machine-learning model that produces images from language descriptions, was the first major effort in this regard.
Initially released in 2021, DALL-E was the first in a wave of text-to-image research built from fundamental advances in language and image processing. The platform was opened to everyone in September 2022.
More recently, Microsoft has also released Bing Image Creator, an AI-powered image generator that incorporates DALL-E to create images based on text prompts.
Other popular AI image generators include Dream by WOMBO, Craiyon, Midjourney, and MyHeritage’s AI Time Machine.
Read More:
- GPT-4 AI Integrated Into Chegg Study Aid Software Despite ChatGPT School Bans
- 16 Best AI Writer Tools for 2023
- Best AI Crypto Tokens & Projects to Invest in 2023
What's the Best Crypto to Buy Now?
- B2C Listed the Top Rated Cryptocurrencies for 2023
- Get Early Access to Presales & Private Sales
- KYC Verified & Audited, Public Teams
- Most Voted for Tokens on CoinSniper
- Upcoming Listings on Exchanges, NFT Drops