According to @DeepLearningAI, researchers from Tencent, Xiamen University, and Fudan University introduced FlashWorld, an AI model that generates high quality, coherent 3D scenes from text or images in seconds, source: @DeepLearningAI. According to @DeepLearningAI, the method combines 2D first and 3D direct pipelines and distills a multi step diffusion process to accelerate text to 3D and image to 3D generation, source: @DeepLearningAI.