Intel Labs, in collaboration with Blockade Labs, has introduced Latent Diffusion Model for 3D (LDM3D), a novel diffusion model that uses generative AI to create realistic 3D visual content.
According to Intel, LDM3D is the industry's first model to generate a depth map using the diffusion process to create 3D images with 360-degree views that are vivid and immersive. LDM3D has the potential to revolutionize content creation, metaverse applications and digital experiences, transforming a wide range of industries, from entertainment and gaming to architecture and design.
"Generative AI technology aims to further augment and enhance human creativity and save time. However, most of today's generative AI models are limited to generating 2D images and only very few can generate 3D images from text prompts," said Vasudev Lal an AI/ML research scientist at Intel Labs. According to Lal, unlike existing latent stable diffusion models, LDM3D allows users to generate an image and a depth map from a given text prompt using almost the same number of parameters. "It provides more accurate relative depth for each pixel in an image compared to standard post-processing methods for depth estimation and saves developers significant time to develop scenes," said Val.
Many of today's advanced generative AI models are limited to generating only 2D images. Unlike existing diffusion models, which generally only generate 2D RGB images from text prompts, LDM3D allows users to generate both an image and a depth map from a given text prompt. Using almost the same number of parameters as latent stable diffusion, LDM3D provides more accurate relative depth for each pixel in an image compared to standard post-processing methods for depth estimation.
LDM3D was trained on a dataset constructed from a subset of 10,000 samples of the LAION-400M database, which contains over 400 million image-caption pairs. The team used the Dense Prediction Transformer (DPT) large-depth estimation model (previously developed at Intel Labs) to annotate the training corpus. The DPT-large model provides highly accurate relative depth for each pixel in an image. The LAION-400M dataset has been built for research purposes to enable testing model training on larger scale for broad researcher and other interested communities.
The LDM3D model is trained on an Intel AI supercomputer powered by Intel Xeon processors and Intel Habana Gaudi AI accelerators. The resulting model and pipeline combine generated RGB image and depth map to generate 360-degree views for immersive experiences.
To demonstrate the potential of LDM3D, Intel and Blockade researchers developed DepthFusion, an application that leverages standard 2D RGB photos and depth maps to create immersive and interactive 360-degree view experiences. DepthFusion utilizes TouchDesigner, a node-based visual programming language for real-time interactive multimedia content, to turn text prompts into interactive and immersive digital experiences. The LDM3D model is a single model to create both an RGB image and its depth map, leading to savings on memory footprint and latency improvements.
LDM3D is being open sourced through HuggingFace. This will allow AI researchers and practitioners to improve this system further and fine-tune it for custom applications.
