Richer Convolutional Features for Edge Detection

Abstract

In this paper, we propose an accurate edge detector using richer convolutional features (RCF). Since objects in natural images possess various scales and aspect ratios, learning the rich hierarchical representations is very critical for edge detection. CNNs have been proved to be effective for this task. In addition, the convolutional features in CNNs gradually become coarser with the increase of the receptive fields. According to these observations, we attempt to adopt richer convolutional features in such a challenging vision task. The proposed network fully exploits multiscale and multilevel information of objects to perform the image-to-image prediction by combining all the meaningful convolutional features in a holistic manner. 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.

A Simple Pooling-Based Design for Real-Time Salient Object Detection

“…These residual blocks are similar to the design in [7] and have channel numbers of {128, 256, 512} from the fine level to the coarse level. As done in [26], each residual block is then followed by a 16channel 3 × 3 convolutional layer for feature compression plus a one-channel 1×1 convolutional layer for edge prediction. We also concatenate these three 16-channel 3 × 3 convolutional layers and feed them to three consecutive 3 × 3 convolutional layers with 48 channels to transmit the captured edge information to the salient object detection branch for detail enhancement.…”

Deep Floor Plan Recognition Using a Multi-Task Network With Room-Boundary-Guided Attention

“…Comparison with a state-of-the-art edge detection network (RCF[12]) on detecting the walls in floor plans.Table 4. A comparison of our full network with Baseline network #1 and Baseline network #2 using the R3D dataset.…”

Salient Building Outline Enhancement and Extraction Using Iterative L0 Smoothing and Line Enhancing

“…To evaluate the performance, we use Canny edge extractor with low and high threshold = (0.05,0.4), L 0 smoothing with higher smoothing level λ = 0.03, OEF [4], Bayesian Salient Edge [5], and RCF [6] as the comparison methods. OEF, Bayesian Salient Edge, and RCF are learning-based methods and use the large object salient outline extraction database BSD500 [16] for training.…”

“…Bayesian Salient Edge [5] based on the OEF and further use Bayesian inference framework to refine the result. Richer convolutional features (RCF) [6] uses multi-stage networks to predict the salient outline.…”