TL;DR: FlashWorld enables fast (7 seconds on a 1x A100/A800 GPU, 4 seconds on 1x H100/H800 GPU) and high-quality 3D scene generation across diverse scenes, from a single image or text prompt.

LLaVA-3D could perform both 2D and 3D vision-language tasks. The left block (b) shows that compared with previous 3D LMMs, our LLaVA-3D achieves state-of-the-art performance across a wide range of 3D ...

Abstract: Accurate segmentation of 3D point clouds in indoor scenes remains a challenging task, often hindered by the labor-intensive nature of data annotation. While weakly supervised learning ...

Abstract: In remote sensing (RS), Few-Shot Novel View Synthesis (FS-NVS) focuses on creating images of unobserved viewpoints using limited training images. Recently, 3D Gaussian Splatting (3DGS) has ...