Search AI Products and News

For a long time, efficiently generating high-quality, wide-angle 3D scenes from a single image has been a challenge faced by researchers. Traditional methods often rely on multi-view data or require time-consuming per-scene optimization, and they struggle with background quality and reconstruction of unseen areas. Existing technologies, when dealing with single-view 3D scene generation, often produce errors or distortions in occluded areas due to insufficient information, along with blurry backgrounds and difficulties in inferring the geometric structure of unseen regions. While regression-based models can synthesize new views in a feedforward manner, they struggle with complex...

The innovative application of Hierarchical 3D Gaussian in virtual reality and computer graphics significantly enhances visual effects, training speed, and real-time rendering capabilities. By constructing a hierarchical 3D Gaussian structure, it efficiently handles large-scale scene datasets, overcoming traditional methods' limitations in sparse scene processing. Utilizing chunked training and hierarchical optimization techniques, it processes te....

最近，来自北京大学和鹏城实验室的研究团队推出了一款名为 HoloDreamer 的新型3D 场景生成框架，旨在通过用户的文本描述生成身临其境、完整的3D 场景。

Recently, top scholars from Tsinghua University and Harvard University have developed a cutting-edge technology called LangSplat. It utilizes three-dimensional Gaussian splatting technology to bring text to life in 3D space, enabling open text queries about the real world.