Unsupervised Adaptation from Repeated Traversals for Autonomous Driving Yurong You 1 Katie Z Luo 1 Travis Zhang

For a self-driving car to operate reliably, its perceptual system must generalize to the end-user's environment -- ideally without additional annotation efforts. One potential solution is to leverage unlabeled data (e.g., unlabeled LiDAR point clouds) collected from the end-users' environments (i.e.