Semantic Image Segmentation with Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform

Abstract

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.

14 pages. Accepted to appear at CVPR 2016

References in corpus (14)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Fully Convolutional Networks for Semantic Segmentation
• Learning Deconvolution Network for Semantic Segmentation
• Fully Connected Deep Structured Networks
• Semantic Image Segmentation via Deep Parsing Network
• BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation
• Attention to Scale: Scale-aware Semantic Image Segmentation
• Pixel-wise Deep Learning for Contour Detection
• Pushing the Boundaries of Boundary Detection using Deep Learning
• Feedforward semantic segmentation with zoom-out features
• DeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour Detection

Cited by in corpus (8)

• 3D fully convolutional networks for subcortical segmentation in MRI: A large-scale study
• Attention to Scale: Scale-aware Semantic Image Segmentation
• Recurrent Neural Networks to Correct Satellite Image Classification Maps
• Learning Affinity via Spatial Propagation Networks
• Fast Semantic Image Segmentation with High Order Context and Guided Filtering
• Face Parsing via Recurrent Propagation
• Superpixel Convolutional Networks using Bilateral Inceptions
• Object Boundary Guided Semantic Segmentation