DeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour Detection

Abstract

Contour detection has been a fundamental component in many image segmentation and object detection systems. Most previous work utilizes low-level features such as texture or saliency to detect contours and then use them as cues for a higher-level task such as object detection. However, we claim that recognizing objects and predicting contours are two mutually related tasks. Contrary to traditional approaches, we show that we can invert the commonly established pipeline: instead of detecting contours with low-level cues for a higher-level recognition task, we exploit object-related features as high-level cues for contour detection. We achieve this goal by means of a multi-scale deep network that consists of five convolutional layers and a bifurcated fully-connected sub-network. The section from the input layer to the fifth convolutional layer is fixed and directly lifted from a pre-trained network optimized over a large-scale object classification task. This section of the network is applied to four different scales of the image input. These four parallel and identical streams are then attached to a bifurcated sub-network consisting of two independently-trained branches. One branch learns to predict the contour likelihood (with a classification objective) whereas the other branch is trained to learn the fraction of human labelers agreeing about the contour presence at a given point (with a regression criterion). We show that without any feature engineering our multi-scale deep learning approach achieves state-of-the-art results in contour detection.

Accepted to CVPR 2015

References in corpus (8)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• DeepPose: Human Pose Estimation via Deep Neural Networks
• OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
• Fully Convolutional Networks for Semantic Segmentation
• DenseNet: Implementing Efficient ConvNet Descriptor Pyramids
• Rich feature hierarchies for accurate object detection and semantic segmentation
• Recurrent Convolutional Neural Networks for Scene Parsing
• PANDA: Pose Aligned Networks for Deep Attribute Modeling

Cited by in corpus (11)

• DenseBox: Unifying Landmark Localization with End to End Object Detection
• Incorporating Network Built-in Priors in Weakly-supervised Semantic Segmentation
• Semantic Image Segmentation with Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform
• Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation
• SemiContour: A Semi-supervised Learning Approach for Contour Detection
• Improved Image Boundaries for Better Video Segmentation
• Weakly Supervised Object Boundaries
• Semantic Segmentation with Boundary Neural Fields
• Convergence rates for pretraining and dropout: Guiding learning parameters using network structure
• Convolutional Channel Features
• Using Cross-Model EgoSupervision to Learn Cooperative Basketball Intention