3D Asset
3D asset generation is a rapidly evolving field focused on creating high-quality, realistic three-dimensional models efficiently and automatically. Current research emphasizes improving the fidelity and speed of generation using various techniques, including diffusion models (often leveraging triplane representations), Gaussian splatting, and multi-view consistency methods. These advancements are significantly impacting fields like gaming, virtual reality, and autonomous driving by providing accessible and efficient tools for 3D content creation, while also raising important considerations around intellectual property protection and copyright.
Papers
3DTopia: Large Text-to-3D Generation Model with Hybrid Diffusion Priors
Fangzhou Hong, Jiaxiang Tang, Ziang Cao, Min Shi, Tong Wu, Zhaoxi Chen, Shuai Yang, Tengfei Wang, Liang Pan, Dahua Lin, Ziwei Liu
ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models
Lukas Höllein, Aljaž Božič, Norman Müller, David Novotny, Hung-Yu Tseng, Christian Richardt, Michael Zollhöfer, Matthias Nießner