Learning to predict crisp boundaries
This addresses the issue of imbalanced data in boundary detection for computer vision applications, offering an incremental improvement over existing methods.
The paper tackles the problem of thick boundary predictions in deep CNNs for edge detection by introducing a novel loss function and an end-to-end network, achieving state-of-the-art results with ODS F-scores of 0.815 on BSDS500 and 0.762 on NYU Depth.
Recent methods for boundary or edge detection built on Deep Convolutional Neural Networks (CNNs) typically suffer from the issue of predicted edges being thick and need post-processing to obtain crisp boundaries. Highly imbalanced categories of boundary versus background in training data is one of main reasons for the above problem. In this work, the aim is to make CNNs produce sharp boundaries without post-processing. We introduce a novel loss for boundary detection, which is very effective for classifying imbalanced data and allows CNNs to produce crisp boundaries. Moreover, we propose an end-to-end network which adopts the bottom-up/top-down architecture to tackle the task. The proposed network effectively leverages hierarchical features and produces pixel-accurate boundary mask, which is critical to reconstruct the edge map. Our experiments illustrate that directly making crisp prediction not only promotes the visual results of CNNs, but also achieves better results against the state-of-the-art on the BSDS500 dataset (ODS F-score of .815) and the NYU Depth dataset (ODS F-score of .762).