Results 261 to 270 of about 5,151,122 (293)

Exploiting Edge Features for Graph Neural Networks [PDF]

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019
Edge features contain important information about graphs. However, current state-of-the-art neural network models designed for graph learning, \eg, graph convolutional networks (GCN) and graph attention networks (GAT), inadequately utilize edge features, especially multi-dimensional edge features. In this paper, we build a new framework for a family of
Liyu Gong, Qiang Cheng 0001
exaly   +3 more sources

Some of the next articles are maybe not open access.

Related searches:

Learning hybrid convolutional features for edge detection

Neurocomputing, 2018
Abstract We present a novel convolutional neural network (CNN) based pipeline which can effectively fuse multi-level information extracted from different intermediate layers generating hybrid convolutional features (HCF) for edge detection. Different from previous methods, the proposed method fuses multi-level information in a feature-map based ...
Xiaowei Hu 0003, Yun Liu 0011, Kai Wang 0001, Bo Ren 0003   +3 more
openaire   +2 more sources

Edge-Based Structural Features

2003
In this chapter we propose edge-based structural features for image representation, develop an algorithm for effective extraction of such features, and explore the effectiveness of these features in aiding visual retrieval tasks [229].
Xiang Sean Zhou, Yong Rui, Thomas S. Huang   +2 more
openaire   +1 more source

Writer identification using edge based features

2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR), 2015
In this paper we present a new method for writer identification, which extract original Local Binary Pattern(LBP) of different radius and Edge descriptors from the edge points of the handwriting. Then, we make combinations of these edge based features.
Zhenyin Fan, Zhenhua Guo 0001, Youbin Chen   +2 more
openaire   +1 more source

An efficient edge-feature constraint visual SLAM

Proceedings of the International Conference on Artificial Intelligence, Information Processing and Cloud Computing, 2019
This paper proposes a constraint method based on the relationship between feature points and object regions. Feature-based visual simultaneous localization and mapping (SLAM) is an effective localization approach for robots in unknown environments.
Wen Liu 0002, Yaokai Mo, Jichao Jiao
openaire   +2 more sources

EGAT: Edge-Featured Graph Attention Network

2021
Most state-of-the-art Graph Neural Networks focus on node features in the learning process but ignore edge features. However, edge features also contain essential information in real-world, such as financial graphs. Node-centric approaches are suboptimal in edge-sensitive graphs since edge features are not adequately utilized.
Ziming Wang, Jun Chen, Haopeng Chen
openaire   +2 more sources

A new edge feature for head-shoulder detection

2013 IEEE International Conference on Image Processing, 2013
In this work, we introduce a new edge feature to improve the head-shoulder detection performance. Since Head-shoulder detection is much vulnerable to vague contour, our new edge feature is designed to extract and enhance the head-shoulder contour and suppress the other contours. The basic idea is that head-shoulder contour can be predicted by filtering
Shu Wang, Jian Zhang 0002, Zhenjiang Miao   +2 more
openaire   +2 more sources

Effect of Transcatheter Edge-to-Edge Repair on Left Ventricular Flow Features

Cardiovascular Engineering and Technology
This study aims to investigate the effects of transcatheter edge-to-edge repair (TEER) on left ventricular hemodynamics and its potential implications for patient health.An in vitro experimental platform was designed to replicate the anatomical and functional characteristics of the left ventricle (LV). This platform integrates native porcine mitral and
Xinyi He, Shuyi Feng, Fan Wu, Hongping Wang, Shizhao Wang, Xiangbin Pan   +5 more
openaire   +2 more sources

Matching with epipolar gradient features and edge transfer

Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429), 2004
A method for quickly and reliably selecting and matching points from three views of a scene is presented. The points that are selected are based on the concept of epipolar gradients, and consist in stable and relevant image features. Then, the selected points are matched using edge transfer, resulting in a measure of consistency for point triplets and ...
Étienne Vincent, Robert Laganière
openaire   +2 more sources

Adjunct Control and edge features

2014
This paper examines a surprising correlation between adjunct control and wh -movement in Portuguese: object control into an adjunct clause may be allowed in addition to subject control if the matrix object undergoes wh -movement.
openaire   +1 more source