Austin R. Waters

Papers

Total Citations

H-Index

About

Austin R. Waters is a researcher pushing the boundaries of embodied AI and 3D scene understanding. His work centers on two critical challenges: enabling robots to follow natural-language instructions in complex environments, and synthesizing photorealistic 3D scenes from limited visual data. In his highly cited 2023 paper, "A New Path: Scaling Vision-and-Language Navigation with Synthetic Instructions and Imitation Learning" (31 citations), Waters tackles the data scarcity problem in Vision-and-Language Navigation (VLN). He introduces a scalable method for generating synthetic navigation instructions and leverages imitation learning to train agents that can robustly follow human commands in photorealistic indoor settings—a key step toward practical, instruction-following robots. Complementing this, his work "Simple and Effective Synthesis of Indoor 3D Scenes" (2023, 19 citations) presents a streamlined approach to generating high-resolution, 3D-consistent images and videos from just one or a few input images, even for viewpoints far beyond the original captures. This research has immediate applications in virtual reality, gaming, and robotics simulation. With a growing citation footprint, Waters is establishing himself as a rising voice in the intersection of computer vision, natural language processing, and 3D graphics.