T-UDA: Temporal Unsupervised Domain Adaptation in Sequential Point Clouds

TL;DR

T-UDA introduces a novel unsupervised domain adaptation method for 3D semantic segmentation in sequential point clouds, combining temporal and cross-sensor geometric consistency with the mean teacher approach, resulting in significant performance improvements.

Contribution

The paper proposes T-UDA, a new domain adaptation technique that leverages temporal and geometric consistency with mean teacher, enhancing 3D segmentation across diverse datasets.

Findings

Massive performance gains on Waymo, nuScenes, SemanticKITTI

Effective across multiple 3D architectures

Publicly available code for reproducibility

Abstract

Deep perception models have to reliably cope with an open-world setting of domain shifts induced by different geographic regions, sensor properties, mounting positions, and several other reasons. Since covering all domains with annotated data is technically intractable due to the endless possible variations, researchers focus on unsupervised domain adaptation (UDA) methods that adapt models trained on one (source) domain with annotations available to another (target) domain for which only unannotated data are available. Current predominant methods either leverage semi-supervised approaches, e.g., teacher-student setup, or exploit privileged data, such as other sensor modalities or temporal data consistency. We introduce a novel domain adaptation method that leverages the best of both trends. Our approach combines input data's temporal and cross-sensor geometric consistency with the mean teacher method. Dubbed T-UDA for "temporal UDA", such a combination yields massive performance gains for the task of 3D semantic segmentation of driving scenes. Experiments are conducted on Waymo Open Dataset, nuScenes and SemanticKITTI, for two popular 3D point cloud architectures, Cylinder3D and MinkowskiNet. Our codes are publicly available at https://github.com/ctu-vras/T-UDA.