Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction

Abstract

Recent works have shown that exploiting multi-scale representations deeply learned via convolutional neural networks (CNN) is of tremendous importance for accurate contour detection. This paper presents a novel approach for predicting contours which advances the state of the art in two fundamental aspects, i.e. multi-scale feature generation and fusion. Different from previous works directly consider- ing multi-scale feature maps obtained from the inner layers of a primary CNN architecture, we introduce a hierarchical deep model which produces more rich and complementary representations. Furthermore, to refine and robustly fuse the representations learned at different scales, the novel Attention-Gated Conditional Random Fields (AG-CRFs) are proposed. The experiments ran on two publicly available datasets (BSDS500 and NYUDv2) demonstrate the effectiveness of the latent AG-CRF model and of the overall hierarchical framework.

Accepted at NIPS 2017

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation
• Holistically-Nested Edge Detection
• Visual Saliency Based on Multiscale Deep Features
• Pushing the Boundaries of Boundary Detection using Deep Learning
• Object Contour Detection with a Fully Convolutional Encoder-Decoder Network
• Crafting GBD-Net for Object Detection
• Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
• SemiContour: A Semi-supervised Learning Approach for Contour Detection

Cited by in corpus (8)

• Richer Convolutional Features for Edge Detection
• PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing
• Lightweight Pixel Difference Networks for Efficient Visual Representation Learning
• Bi-Directional Cascade Network for Perceptual Edge Detection
• Learning Deep Morphological Networks with Neural Architecture Search
• Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
• Learning Crisp Boundaries Using Deep Refinement Network and Adaptive Weighting Loss
• SketchEdit: Mask-Free Local Image Manipulation with Partial Sketches