Integrating parametric and non-parametric models for scene labeling

By Bing Shuai, G. Wang, Zhen Zuo, Bing Wang, Lifan Zhao. 2015

53 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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. G. Wang
2. Jianfei Cai
3. Jiuxiang Gu
4. Haibin Ling
5. Amir Shahroudy
6. Yi Yang
7. Jun Liu
8. Heng Fan

In Media Technology

1. G. Wang
2. Heng Fan
3. Zhen Zuo
4. Jianfei Cai
5. Haibin Ling
6. Zhenhua Wang
7. Gang Wang
8. Xingxing Wang

Citations in the corpus

• Recent advances in convolutional neural networks (2015, background): “problems, and [188] solve it by integrating global beliefs.”
• DAG-Recurrent Neural Networks for Scene Labeling (2015, background): “Shuai et al.[21] models the global-order dependencies in a non-parametric framework to disambiguate the local confusions.”
• Learning fine-grained features via a CNN Tree for Large-scale Classification (2015, background): “Convolutional neural networks (CNN) [1], [2] have recently demonstrated superior performance on many tasks such as image classification [3], [4], [5], object detection [6], [7], [8], [9], [10], [11], object tracking [12], [13], [14], text detection [15], [16], text recognition [17], [18], [19], vide”
• Gated Siamese Convolutional Neural Network Architecture for Human Re-identification (2016, background): “Deep Learning for Human Re-Identification: Convolutional Neural Networks have achieved phenomenal results on several computer vision tasks [13,37, 38,42].”
• Dense Recurrent Neural Networks for Scene Labeling (2018, background): “e unary and pairwise features in a 2nd-order Conditional Random Field (CRF). Zhang et al.[34] and Roy et al. [18] build a fully connected graph to enforce higher order labeling coherence. Shuai et al.[21] models the global-order dependencies in a nonparametric framework to disambiguate the local con”
• Quaddirectional 2D-Recurrent Neural Networks For Image Labeling (2015, background): “Recently, deep neural networks [1]–[3] are shown that they can learn robust features to differentiate the visual subtleties of local patches.”
• Region-Based Semantic Segmentation with End-to-End Training (2016, unknown): “Pixel-level class frequencies are often unbalanced [2, 9, 11, 12, 14, 16, 20, 21, 40– 42].”