Shulou(Shulou.com)11/24 Report--

CTOnews.com Intel Research today announced a partnership with Blockade Labs to release the Latent Diffusion Model for 3D (LDM3D), a new diffusion model that uses generative AI to create 3D visual content.

LDM3D is an industry-leading model that generates depth maps using diffusion processes to generate realistic, immersive 360-degree panoramic images. LDM3D promises to revolutionize content creation, meta-cosmic applications and digital experiences, transforming many industries including entertainment, gaming, architecture and design.

Intel says LDM3D was trained on a subset of 10,000 samples from the LAION-400M dataset. LAION-400M is a large graphic-text dataset containing more than 400 million graphic-text pairs. When labeling the training corpus, the research team used DPT-Large, a dense depth estimation model previously developed by Intel Research, which provides a highly accurate relative depth for each pixel in the image. The LAION-400M dataset was created for research purposes so that researchers and other interested communities could test model training on a larger scale.

The LDM3D model was trained on an Intel AI supercomputer powered by Intel Xeon processors and Intel Habana Gaudi AI accelerators. The final model and process integrates RGB images and depth maps to produce a 360-degree panorama for an immersive experience.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.