Richer Convolutional Features for Edge Detection

By Yun Liu, Yun Liu, Ming-Ming Cheng, Xiaowei Hu, Kai Wang, Xiang Bai. 2016

1,215 citations recorded by the source corpus.

Edge detection is a fundamental problem in computer vision. Recently, convolutional neural networks (CNNs) have pushed forward this field significantly. Existing methods which adopt specific layers of deep CNNs may fail to capture complex data structures caused by variations of scales and aspect ratios. In this paper, we propose an accurate edge detector using richer convolutional features (RCF). RCF encapsulates all convolutional features into more discriminative representation, which makes good usage of rich feature hierarchies, and is amenable to training via backpropagation. RCF fully exploits multiscale and multilevel information of objects to perform the image-to-image prediction holistically. Using VGG16 network, we achieve state-of-the-art performance on several available datasets. When evaluating on the well-known BSDS500 benchmark, we achieve ODS F-measure of 0.811 while retaining a fast speed (8 FPS). Besides, our fast version of RCF achieves ODS F-measure of 0.806 with 30 FPS. We also demonstrate the versatility of the proposed method by applying RCF edges for classical image segmentation.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning (2019, background): “There is a large body of literature that addresses “Image-to-Edges” problems [5, 10, 26, 29].”
Res2Net: A New Multi-Scale Backbone Architecture (2019, methodology): “Unsurprisingly, multi-scale features have been widely used in both conventional feature design [1], [31] and deep learning [6], [18], [29], [39], [44], [52].”
Deepstrip: High-Resolution Boundary Refinement (2020, background): “Most research in boundary detection focuses on improving the boundary quality in LR through introducing more semantic information [2, 44, 26] or human interaction [22, 25, 42, 24, 7].”
Bi-Directional Cascade Network for Perceptual Edge Detection (2019, methodology): “Multiple scale cues are also used in many approaches [48, 39, 24, 50, 30, 34, 51].”
EdgeConnect: Structure Guided Image Inpainting using Edge Prediction (2019, background): “There is a large body of literature that addresses “Image-to-Edges” problems [6, 11, 29, 32].”
Instance-level Human Parsing via Part Grouping Network (2018, methodology): “To evaluate instance-aware edge detection performance, we use the same measures for traditional edge detection [27]: ﬁxed contour threshold (ODS) and per-image best threshold (OIS).”
Cali-Sketch: Stroke Calibration and Completion for High-Quality Face Image Generation from Poorly-Drawn Sketches (2019, background): “Contours can be obtained by the boundary contour edge extractors like HED [45], COB [59], [60], RCF [61], [62], or similar to pix2pixHD [5], simpliﬁed from the face parsing semantic labels.”
A Survey of Deep Learning-Based Object Detection (2019, background): “[264] present an accurate edge detector which utilizes richer convolutional features.”
SegFix: Model-Agnostic Boundary Refinement for Segmentation (2020, background): “Boundary for Segmentation: Some previous efforts [1,56,42,57] focused on localizing semantic boundaries.”
Learning a Discriminative Feature Network for Semantic Segmentation (2018, methodology): “There are lots of specific methods proposed for the task of boundary detection [39, 36, 37, 25].”
Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction (2017, methodology): “We also report results of the RCF method [23], although they are not comparable because in [23] an extra dataset (Pascal Context) was used during RCF training to improve the results on BSDS500.”
Weakly-Supervised Salient Object Detection via Scribble Annotations (2020, methodology): “2 from RCF edge detection network [22] to train EDN.”
Interactive Two-Stream Decoder for Accurate and Fast Saliency Detection (2020, background): “For other researches please refer the recent survey [2, 6, 34, 23, 33] for more details.”
Learning to predict crisp boundaries (2018, methodology): “We compare against the state-of-the-art methods which include OEF [49], gPb-UCM [5], gPb+NG [50], SE [26], SE+NG+ [51], HED [12] and RCF [13].”
Deep Structural Contour Detection (2020, methodology): “RCF [27] upgrades the skip-layer structure of HED by making independent predictions from each convolutional layer of the VGG model and achieves better results.”