Distantly Supervised Road Segmentation
This reduces annotation costs for road segmentation tasks, but it is incremental as it adapts existing methods to a new supervision paradigm.
The paper tackles road segmentation with minimal annotation by using distantly supervised image-level labels to generate weak pixel-wise masks, achieving 93.8% of the performance of a fully supervised approach on Cityscapes.
We present an approach for road segmentation that only requires image-level annotations at training time. We leverage distant supervision, which allows us to train our model using images that are different from the target domain. Using large publicly available image databases as distant supervisors, we develop a simple method to automatically generate weak pixel-wise road masks. These are used to iteratively train a fully convolutional neural network, which produces our final segmentation model. We evaluate our method on the Cityscapes dataset, where we compare it with a fully supervised approach. Further, we discuss the trade-off between annotation cost and performance. Overall, our distantly supervised approach achieves 93.8% of the performance of the fully supervised approach, while using orders of magnitude less annotation work.