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

By Dan Xu, Wanli Ouyang, Xavier Alameda-Pineda, Elisa Ricci, Xiaogang Wang, Nicu Sebe. 2017

134 citations recorded by the source corpus.

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.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Model Architectures

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

1. Luc Van Gool
2. Xiaogang Wang
3. Ming-Ming Cheng
4. Ming-Hsuan Yang
5. Yi Yang
6. Wanli Ouyang
7. Dacheng Tao
8. Ling Shao

In Media Technology

1. Nicu Sebe
2. Dan Xu
3. Wanli Ouyang
4. Elisa Ricci
5. Haibin Ling
6. Xavier Alameda-Pineda
7. Xiang Bai
8. Yaping Huang

In Model Architectures

1. Wanli Ouyang
2. Xavier Alameda-Pineda
3. Dan Xu
4. Xiang Bai
5. Nicu Sebe
6. Yaping Huang
7. Shiliang Zhang
8. Haibin Ling

Citations in the corpus

• EDTER: Edge Detection with Transformer (2022, methodology): “[49], gPb+NG [19], SE [15], SE+NG+ [20], OEF [21], SemiContour [69], HED [65], RCF [36], AMH-Net [66], LPCB [12], BDCN [22], and PiDiNet [53].”
• RINDNet: Edge Detection for Discontinuity in Reflectance, Illumination, Normal and Depth (2021, background): “CNN-based edge detectors [3, 4, 9, 10, 17, 20, 24, 26, 31, 36, 45] do not rely on hand-crafted features and achieve better performance.”
• Every Smile is Unique: Landmark-Guided Diverse Smile Generation (2018, background): “attention models or structured multiscale features [48]) of high-level subjective properties [1] in facial expression generation.”
• PAD-Net: Multi-tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing (2018, background): “Recently, researchers studied implementing the CRF models with CNN enabling the end-to-end optimization of the whole deep network [32, 54, 53].”
• Bi-Directional Cascade Network for Perceptual Edge Detection (2019, methodology): “Note that, AMH-Net applies the deeper ResNet50 to construct the edge detector.”
• Richer Convolutional Features for Edge Detection (2016, background): “and some recent deep learning based approaches, including DeepContour [15], DeepEdge [13], HED [16], HFL [35], MIL+G-DSN+MS +NCuts [32], CASENet [29], AMH [28], CED [27] and etc.”