Semantic Image Segmentation with Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform
Liang-Chieh Chen, Jonathan T. Barron, George Papandreou et al. · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · 2016
Deep convolutional neural networks (CNNs) are the backbone of state-of-art semantic image segmentation systems. Recent work has shown that complementing CNNs with fully-connected conditional random fields (CRFs) can significantly enhance their object localization accuracy, yet dense CRF inference is computationally expensive. We propose replacing the fully-connected CRF with domain transform (DT), a modern edge-preserving filtering method in which the amount of smoothing is controlled by a reference edge map. Domain transform filtering is several times faster than dense CRF inference and we show that it yields comparable semantic segmentation results, accurately capturing object boundaries. Importantly, our formulation allows learning the reference edge map from intermediate CNN features instead of using the image gradient magnitude as in standard DT filtering. This produces task-specific edges in an end-to-end trainable system optimizing the target semantic segmentation quality.
Cited by 364 works. doi:10.1109/cvpr.2016.492 Open-access PDF
Authors
References
- Long Short-Term Memory
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- ImageNet Large Scale Visual Recognition Challenge
- Backpropagation Applied to Handwritten Zip Code Recognition
- Fully convolutional networks for semantic segmentation
- Microsoft COCO: Common Objects in Context
- Caffe
- Finding Structure in Time
- The Pascal Visual Object Classes Challenge: A Retrospective
- Hypercolumns for object segmentation and fine-grained localization