Deep Hough-Transform Line Priors
This work addresses the need for more data-efficient methods in computer vision for line detection, though it is incremental as it builds on classical knowledge-based priors.
The paper tackles the problem of reducing dependency on labeled data for line segment detection by integrating a trainable Hough transform block into a deep network, which improves data efficiency by leveraging prior knowledge about global line parameterizations.
Classical work on line segment detection is knowledge-based; it uses carefully designed geometric priors using either image gradients, pixel groupings, or Hough transform variants. Instead, current deep learning methods do away with all prior knowledge and replace priors by training deep networks on large manually annotated datasets. Here, we reduce the dependency on labeled data by building on the classic knowledge-based priors while using deep networks to learn features. We add line priors through a trainable Hough transform block into a deep network. Hough transform provides the prior knowledge about global line parameterizations, while the convolutional layers can learn the local gradient-like line features. On the Wireframe (ShanghaiTech) and York Urban datasets we show that adding prior knowledge improves data efficiency as line priors no longer need to be learned from data. Keywords: Hough transform; global line prior, line segment detection.