Image Source: Generated by Wujie AI

Developer Stability AI announced several new powerful features for its Stable Diffusion platform today. These updates not only provide exciting new text-to-image capabilities, but also venture into the realm of 3D content creation.

3D Model Generation Arrives

The most significant enhancement is the all-new Stable 3D model. Up to this point, Stable Diffusion has primarily focused on 2D image generation. The Stable 3D model will change that by providing capabilities for 3D content creation across various types, including graphic design and even video game development.

For graphic designers, digital artists, and game developers, 3D content creation can be one of the most complex and time-consuming tasks, often taking hours (sometimes even days) to create a moderately complex 3D object.

Stable 3D is able to generate concept-quality textured 3D objects through images or illustrations, or by writing text prompts, eliminating much of the complexity and allowing non-experts to generate draft-quality 3D models in minutes by selecting a model.

Objects created using Stable 3D are in the ".obj" standard file format, which can be further edited and improved in 3D tools such as Blender and Maya, or imported into game engines such as Unreal Engine 5 or Unity, significantly reducing the workload for creators.

Stable 3D provides a fast creative environment for independent designers, artists, and developers, enabling them to create thousands of 3D objects at extremely low cost every day.

Currently, Stability AI has only opened a private preview version of Stable 3D, and applications can be made at https://stability.ai/contact.

Sky Replacer

In addition to entering the 3D content generation field, Stability AI has also launched the Sky Replacer tool, designed to do exactly as the name suggests—replace the sky background in 2D images.

The Stable Diffusion platform now also offers Stable Fine-Tuning, aimed at helping businesses expedite the image fine-tuning process for specific use cases.

Furthermore, the company will integrate an invisible watermark for content authentication into images generated by the Stability AI API. As generative AI becomes an increasingly common part of workflows, these new updates are aimed at helping businesses in creative development.

Stability AI CEO Emad Mostaque stated in an interview, "This is about giving creative storytellers the tools they need to have additional control over their images."

Increasingly Competitive AIGC Field

Stability AI's advancements come at a time when the market for text-to-image generation is becoming increasingly competitive.

Adobe has set its sights on this market, launching AI tool Firefly tightly integrated with its design software. Midjourney continues to add new features to its technology to help designers generate images. Notably, OpenAI recently released the native DALL-E 3 model for ChatGPT, which improves the ability to generate text within images.

Mostaque is acutely aware of his competitive position and is committed to helping Stability AI stand out in multiple ways. He emphasizes that his company is now moving from just providing models to providing creative channels. He points out that with the new Sky Replacer and fine-tuning features, they are going beyond the core basic models used for image generation with additional steps.

Sky Replacer is not just a feature; it is focused on business use cases.

The concept of replacing backgrounds in images is not a new one. In non-generative AI applications, backgrounds can often be replaced using techniques such as green screens and chroma keying.

Mostaque states that Stability AI is building workflows on top of these classic techniques and automating workflows to make business users' processes faster and more efficient. Changing the background of the sky is not just about adding some form of creative flair; it is a feature with very specific and practical use cases.

"For example, Sky Replacer is very useful for real estate."

Mostaque points out that users want to have different backgrounds and different lighting effects. He emphasizes that fundamentally, all of this is about providing control, as organizations have their own workflows for generating images and content. What Stability AI is doing is building optimized workflows to help achieve the control needed for different use cases.

"Sky Replacer is the first in a series of products we will be rolling out, which are very industry and enterprise-focused, built on our experience over the past 6 to 12 months," he said.

Stable 3D Expands New Use Cases for Stable Diffusion

The new Stable 3D model works by expanding the diffusion model used in Stable Diffusion to include additional 3D datasets and vectorization.

"I'm very excited about being able to create entire worlds in 3D,"

Mostaque explains that Stable 3D is built on the work of Stable Diffusion and Stability AI on Objaverse-XL, one of the world's largest open 3D datasets. Building and rendering 3D images has long been a resource-intensive process, but Mostaque is optimistic that Stable 3D will be more efficient than traditional 3D image generation methods. He emphasizes that stable AI is still in its early stages, but he is optimistic that the technology will steadily develop and expand over time. Stable 3D is initially available in a private preview.

"This is very efficient compared to traditional 3D model creation. What used to take a long time to build can now be approved very quickly,"

Watermarking and Biden's Executive Order on AI

The Biden administration issued an executive order (EO) on artificial intelligence this week, one component of which is integrating watermarks into generated content.

Stability AI is now integrating invisible watermarks and content credentials into its API. Content credentials are part of a multi-vendor industry effort involving Adobe and other companies aimed at providing information about the authors of content. Mostaque states that adding invisible watermarks and content credentials is a responsible practice. This is also part of a broader effort by Stability AI to bring authenticity to generated content.

Mostaque said, "We are indeed launching a series of measures and some additional research around this issue, because we want to know what is real and what is fake. It also helps us build some attribution and other mechanisms for future versions."

References:

https://venturebeat.com/ai/exclusive-stability-ai-brings-advanced-3d-and-image-fine-tuning-to-stable-diffusion/

https://stability.ai/blog/stability-ai-enhanced-image-apis-for-business-features

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。