Talk

Infinite Photorealistic Worlds using Procedural Generation

avatar
Zeyu Ma
2023/8/21
Procedural Generation
3D Content Generation
main image for this paper

Abstract

Randomly generated, non cherry-picked images produced by Infinigen. From top left to bottom right: Forest, River, Underwater, Caves, Coast, Desert, Mountain and Plains. See Appendix for more samples.

Speaker

Zeyu Ma is a fourth-year PhD candidate in the Vision & Learning Lab at Princeton University, where he is supervised by Prof. Jia Deng. He received his B.Eng. in Electronic Engineering from Tsinghua University in 2020. Ma’s current research focuses on multi-view 3D reconstruction and procedural data generation for 3D vision tasks.

Video

Extra Details

Speaker Website / Paper Link / Paper Project Page / Paper Code