Richer Convolutional Features for Edge Detection
arXiv:1612.02103 · doi:10.1109/TPAMI.2018.2878849
Abstract
In this paper, we propose an accurate edge detector using richer convolutional features (RCF). Since objects in nature images have various scales and aspect ratios, the automatically learned rich hierarchical representations by CNNs are very critical and effective to detect edges and object boundaries. And the convolutional features gradually become coarser with receptive fields increasing. Based on these observations, our proposed network architecture makes full use of multiscale and multi-level information to perform the image-to-image edge prediction by combining all of the useful convolutional features into a holistic framework. It is the first attempt to adopt such rich convolutional features in computer vision tasks. Using VGG16 network, we achieve \sArt results on several available datasets. When evaluating on the well-known BSDS500 benchmark, we achieve ODS F-measure of \textbf{.811} while retaining a fast speed (\textbf{8} FPS). Besides, our fast version of RCF achieves ODS F-measure of \textbf{.806} with \textbf{30} FPS.
References in corpus (10)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Caffe: Convolutional Architecture for Fast Feature Embedding
- R-FCN: Object Detection via Region-based Fully Convolutional Networks
- Fully Convolutional Networks for Semantic Segmentation
- DenseNet: Implementing Efficient ConvNet Descriptor Pyramids
- Convolutional Oriented Boundaries
- Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction
- Deep Crisp Boundaries: From Boundaries to Higher-level Tasks
- Pixel-wise Deep Learning for Contour Detection
- WebSeg: Learning Semantic Segmentation from Web Searches
Cited by in corpus (114)
- Res2Net: A New Multi-scale Backbone Architecture
- A Survey of Deep Learning-based Object Detection
- Deeply supervised salient object detection with short connections
- EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
- STAR: A Structure and Texture Aware Retinex Model
- CrackGAN: Pavement Crack Detection Using Partially Accurate Ground Truths Based on Generative Adversarial Learning
- HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline
- Learning Depth from Single Images with Deep Neural Network Embedding Focal Length
- Deep Hough Transform for Semantic Line Detection
- DFR: Deep Feature Reconstruction for Unsupervised Anomaly Segmentation
- Dense Extreme Inception Network for Edge Detection
- Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge and Skeleton
- PP-LCNet: A Lightweight CPU Convolutional Neural Network
- Learning a Discriminative Feature Network for Semantic Segmentation
- Cross-SRN: Structure-Preserving Super-Resolution Network with Cross Convolution
- Confusing Image Quality Assessment: Towards Better Augmented Reality Experience
- A Neural Network based Shock Detection and Localization Approach for Discontinuous Galerkin Methods
- Feature Pyramid and Hierarchical Boosting Network for Pavement Crack Detection
- Free-form tumor synthesis in computed tomography images via richer generative adversarial network
- Lightweight Pixel Difference Networks for Efficient Visual Representation Learning
- Fast and Accurate Road Crack Detection Based on Adaptive Cost-Sensitive Loss Function
- Hierarchical Convolutional Neural Network with Feature Preservation and Autotuned Thresholding for Crack Detection
- Distance transform regression for spatially-aware deep semantic segmentation
- Object Contour and Edge Detection with RefineContourNet
- MiniSeg: An Extremely Minimum Network for Efficient COVID-19 Segmentation
- Semantic Edge Detection with Diverse Deep Supervision
- Deep Ice Layer Tracking and Thickness Estimation using Fully Convolutional Networks
- Learning Regional Attraction for Line Segment Detection
- Regularized Densely-connected Pyramid Network for Salient Instance Segmentation
- Bi-Directional Cascade Network for Perceptual Edge Detection
- Learning Deep Morphological Networks with Neural Architecture Search
- HS-ResNet: Hierarchical-Split Block on Convolutional Neural Network
- Revisiting Computer-Aided Tuberculosis Diagnosis
- Instance-level Human Parsing via Part Grouping Network
- Local Label Point Correction for Edge Detection of Overlapping Cervical Cells
- Graph Jigsaw Learning for Cartoon Face Recognition
- Adaptive Linear Span Network for Object Skeleton Detection
- Improving Camouflaged Object Detection with the Uncertainty of Pseudo-edge Labels
- Cali-Sketch: Stroke Calibration and Completion for High-Quality Face Image Generation from Human-Like Sketches
- EdgeStereo: A Context Integrated Residual Pyramid Network for Stereo Matching
- Residual Pyramid Learning for Single-Shot Semantic Segmentation
- Weakly-Supervised Salient Object Detection via Scribble Annotations
- DNA: Deeply-supervised Nonlinear Aggregation for Salient Object Detection
- Semi-Supervised Self-Taught Deep Learning for Finger Bones Segmentation
- Deep Floor Plan Recognition Using a Multi-Task Network with Room-Boundary-Guided Attention
- TTPLA: An Aerial-Image Dataset for Detection and Segmentation of Transmission Towers and Power Lines
- A relic sketch extraction framework based on detail-aware hierarchical deep network
- Does enhanced shape bias improve neural network robustness to common corruptions?
- Pixel Difference Networks for Efficient Edge Detection
- Rethinking Interactive Image Segmentation: Feature Space Annotation
- SREdgeNet: Edge Enhanced Single Image Super Resolution using Dense Edge Detection Network and Feature Merge Network
- SegFix: Model-Agnostic Boundary Refinement for Segmentation
- EdgeStereo: An Effective Multi-Task Learning Network for Stereo Matching and Edge Detection
- SE2Net: Siamese Edge-Enhancement Network for Salient Object Detection
- 3D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Prior
- Learning to Calibrate Straight Lines for Fisheye Image Rectification
- Fisheye Distortion Rectification from Deep Straight Lines
- Beef Cattle Instance Segmentation Using Fully Convolutional Neural Network
- Unsupervised Medical Image Segmentation with Adversarial Networks: From Edge Diagrams to Segmentation Maps
- Optimizing the F-measure for Threshold-free Salient Object Detection
- Three Birds One Stone: A General Architecture for Salient Object Segmentation, Edge Detection and Skeleton Extraction
- Recent Advances in the Applications of Convolutional Neural Networks to Medical Image Contour Detection
- Shearlets as Feature Extractor for Semantic Edge Detection: The Model-Based and Data-Driven Realm
- Visualized Insights into the Optimization Landscape of Fully Convolutional Networks
- WebSeg: Learning Semantic Segmentation from Web Searches
- Photo-Sketching: Inferring Contour Drawings from Images
- Dark Side Augmentation: Generating Diverse Night Examples for Metric Learning
- Panoptic Edge Detection
- Salient Object Detection via High-to-Low Hierarchical Context Aggregation
- Sketch2code: Generating a website from a paper mockup
- Dynamic Feature Fusion for Semantic Edge Detection
- Unsupervised Pixel-level Road Defect Detection via Adversarial Image-to-Frequency Transform
- Learning Deep Models from Synthetic Data for Extracting Dolphin Whistle Contours
- SLURP: Side Learning Uncertainty for Regression Problems
- Unmixing Convolutional Features for Crisp Edge Detection
- A Simple Pooling-Based Design for Real-Time Salient Object Detection
- Towards Label-Free 3D Segmentation of Optical Coherence Tomography Images of the Optic Nerve Head Using Deep Learning
- Applications of Deep Learning in Fundus Images: A Review
- A Unified Efficient Pyramid Transformer for Semantic Segmentation
- Deep Green Function Convolution for Improving Saliency in Convolutional Neural Networks
- CAFENet: Class-Agnostic Few-Shot Edge Detection Network
- DeepFlux for Skeletons in the Wild
- Traditional Method Inspired Deep Neural Network for Edge Detection
- Learnable Gabor modulated complex-valued networks for orientation robustness
- TricycleGAN: Unsupervised Image Synthesis and Segmentation Based on Shape Priors
- FINED: Fast Inference Network for Edge Detection
- Skip-WaveNet: A Wavelet based Multi-scale Architecture to Trace Snow Layers in Radar Echograms
- Compact Twice Fusion Network for Edge Detection
- Robust Regression via Deep Negative Correlation Learning
- Fast and Optimal Laplacian Solver for Gradient-Domain Image Editing using Green Function Convolution
- Weak Edge Identification Nets for Ocean Front Detection
- DeepStrip: High Resolution Boundary Refinement
- Saliency Enhancement using Gradient Domain Edges Merging
- RED-NET: A Recursive Encoder-Decoder Network for Edge Detection
- TP-LSD: Tri-Points Based Line Segment Detector
- Dense Prediction with Attentive Feature Aggregation
- DOOBNet: Deep Object Occlusion Boundary Detection from an Image
- Learning Crisp Edge Detector Using Logical Refinement Network
- A new baseline for edge detection: Make Encoder-Decoder great again
- Rethinking the Image Feature Biases Exhibited by Deep CNN Models
- Hi-Fi: Hierarchical Feature Integration for Skeleton Detection
- ProMask: Probability Mask for Skeleton Detection
- Improving Convolutional Neural Networks Via Conservative Field Regularisation and Integration
- DDNet: Dual-path Decoder Network for Occlusion Relationship Reasoning
- Salient Building Outline Enhancement and Extraction Using Iterative L0 Smoothing and Line Enhancing
- Topology Maintained Structure Encoding
- Learning Pixel-wise Labeling from the Internet without Human Interaction
- Multi-Scale Tensorial Summation and Dimensional Reduction Guided Neural Network for Edge Detection
- Contour Primitive of Interest Extraction Network Based on One-Shot Learning for Object-Agnostic Vision Measurement
- Learning Aesthetic Layouts via Visual Guidance
- ELKPPNet: An Edge-aware Neural Network with Large Kernel Pyramid Pooling for Learning Discriminative Features in Semantic Segmentation
- On segmentation of pectoralis muscle in digital mammograms by means of deep learning
- TextContourNet: a Flexible and Effective Framework for Improving Scene Text Detection Architecture with a Multi-task Cascade
- Low-Light Maritime Image Enhancement with Regularized Illumination Optimization and Deep Noise Suppression