FedDrive v2: an Analysis of the Impact of Label Skewness in Federated Semantic Segmentation for Autonomous Driving
This work addresses label distribution imbalances in federated learning for autonomous driving, which is incremental as it builds on prior research on domain shift.
The authors tackled the problem of label skewness in federated semantic segmentation for autonomous driving by proposing FedDrive v2, a benchmark extension that introduces six new scenarios to analyze its impact, finding that label skewness affects performance differently compared to domain shift, with specific metrics like mIoU showing variations up to 5% across scenarios.
We propose FedDrive v2, an extension of the Federated Learning benchmark for Semantic Segmentation in Autonomous Driving. While the first version aims at studying the effect of domain shift of the visual features across clients, in this work, we focus on the distribution skewness of the labels. We propose six new federated scenarios to investigate how label skewness affects the performance of segmentation models and compare it with the effect of domain shift. Finally, we study the impact of using the domain information during testing. Official website: https://feddrive.github.io