CASENet: Deep Category-Aware Semantic Edge Detection
This work addresses semantic edge detection for computer vision tasks, offering a significant improvement over existing methods, though it is incremental as it builds on deep learning approaches.
The paper tackles the challenging multi-label problem of category-aware semantic edge detection, where edge pixels can belong to multiple classes, and proposes a novel deep learning architecture based on ResNet with a skip-layer design and multi-label loss, achieving state-of-the-art performance by a large margin on datasets like SBD and Cityscapes.
Boundary and edge cues are highly beneficial in improving a wide variety of vision tasks such as semantic segmentation, object recognition, stereo, and object proposal generation. Recently, the problem of edge detection has been revisited and significant progress has been made with deep learning. While classical edge detection is a challenging binary problem in itself, the category-aware semantic edge detection by nature is an even more challenging multi-label problem. We model the problem such that each edge pixel can be associated with more than one class as they appear in contours or junctions belonging to two or more semantic classes. To this end, we propose a novel end-to-end deep semantic edge learning architecture based on ResNet and a new skip-layer architecture where category-wise edge activations at the top convolution layer share and are fused with the same set of bottom layer features. We then propose a multi-label loss function to supervise the fused activations. We show that our proposed architecture benefits this problem with better performance, and we outperform the current state-of-the-art semantic edge detection methods by a large margin on standard data sets such as SBD and Cityscapes.