A Cross-Season Correspondence Dataset for Robust Semantic Segmentation
This work addresses robustness in semantic segmentation for applications like autonomous driving or mapping, but it is incremental as it builds on existing methods with new datasets.
The paper tackles the problem of semantic segmentation robustness to seasonal and weather changes by using 2D-2D point matches across images to enforce label consistency during training, resulting in improved segmentation performance.
In this paper, we present a method to utilize 2D-2D point matches between images taken during different image conditions to train a convolutional neural network for semantic segmentation. Enforcing label consistency across the matches makes the final segmentation algorithm robust to seasonal changes. We describe how these 2D-2D matches can be generated with little human interaction by geometrically matching points from 3D models built from images. Two cross-season correspondence datasets are created providing 2D-2D matches across seasonal changes as well as from day to night. The datasets are made publicly available to facilitate further research. We show that adding the correspondences as extra supervision during training improves the segmentation performance of the convolutional neural network, making it more robust to seasonal changes and weather conditions.