Joint Segmentation and Path Classification of Curvilinear Structures
This work addresses the challenge of inferring graph representations for curvilinear networks, which is important for applications like road mapping and neuron analysis, but it appears incremental as it combines existing tasks into a single pipeline.
The paper tackles the problem of detecting curvilinear structures in images by jointly performing segmentation and path classification with a deep network, showing that this approach enforces consistency across the pipeline and is applied on roads and neurons datasets.
Detection of curvilinear structures in images has long been of interest. One of the most challenging aspects of this problem is inferring the graph representation of the curvilinear network. Most existing delineation approaches first perform binary segmentation of the image and then refine it using either a set of hand-designed heuristics or a separate classifier that assigns likelihood to paths extracted from the pixel-wise prediction. In our work, we bridge the gap between segmentation and path classification by training a deep network that performs those two tasks simultaneously. We show that this approach is beneficial because it enforces consistency across the whole processing pipeline. We apply our approach on roads and neurons datasets.