The Neverwhere Visual Parkour Benchmark Suite

Ziyu Chen, Haoran Chang, Henghui Bao, Ran Choi, Alan Yu, Ri-Zhao Qiu, Yajvan Ravan, John J. Leonard, Xiaolong Wang, Phillip Isola, Yue Wang, Ge Yang

Keywords: parkour, locomotion, real-to-sim-to-real, gaussian splatting

Abstract: State-of-the-art visual locomotion controllers are increasingly capable at handling complex visual environments, making evaluating their real-world performance before deployment increasingly difficult. This work intends to narrow this train/evaluation gap by developing a collection of hyper-photo-realistic, closed-loop evaluation environments -- The Neverwhere Benchmark Suite -- comprised of over sixty 3D Gaussian Splatting of urban indoor and outdoor scenes. Our goal is to encourage large-scale and reproducible robot evaluation by making it easier to create and integrate Gaussian splats-based reconstructions into simulated continuous testing setups. We also underscore the potential pitfalls of relying exclusively on 3D Gaussian-generated data for training, by providing policy checkpoints trained over multiple Neverwhere scenes and their performance when evaluated in novel scenes. Our analysis illustrates the necessity of sourcing diverse data to ensure performance.

Dataset URL: https://www.dropbox.com/scl/fo/jjqtoxtmt7uvu7v1f6mtf/AAY3gp55UtQ9eYMqjudyU2U?rlkey=50etjwsz8c8xomb3rvw7tmjb0&st=iziabk1n&dl=0

Code URL: https://anonymoususer362.github.io/neverwhere/

Primary Area: Data for Reinforcement learning (e.g., decision and control, planning, hierarchical RL, robotics)

Submission Number: 362