Domain Adaptation in LiDAR Semantic Segmentation via Alternating Skip Connections and Hybrid Learning
This addresses the problem of domain shift for researchers and practitioners in autonomous driving and robotics, though it appears incremental as it builds on existing GAN and segmentation techniques.
The paper tackles domain adaptation in LiDAR semantic segmentation by proposing a hybrid framework that combines GAN-based image-to-image translation with alternating skip connections and a state-of-the-art segmentation network, achieving superior performance compared to baselines and prior methods on benchmark datasets.
In this paper we address the challenging problem of domain adaptation in LiDAR semantic segmentation. We consider the setting where we have a fully-labeled data set from source domain and a target domain with a few labeled and many unlabeled examples. We propose a domain adaption framework that mitigates the issue of domain shift and produces appealing performance on the target domain. To this end, we develop a GAN-based image-to-image translation engine that has generators with alternating connections, and couple it with a state-of-the-art LiDAR semantic segmentation network. Our framework is hybrid in nature in the sense that our model learning is composed of self-supervision, semi-supervision and unsupervised learning. Extensive experiments on benchmark LiDAR semantic segmentation data sets demonstrate that our method achieves superior performance in comparison to strong baselines and prior arts.