Patent
Image processing method and device, storage medium, computer program and electronic equipment
Yang Wei,Ouyang Wanli,Li Shuang,Li Hongsheng,Wang Xiaogang +4 more
- 29 Jun 2018
14
TL;DR: In this paper, the image processing method comprises the following stepsthat: obtaining the characteristic pattern of an image to be detected; through a neural network, on the basis of at least two different scales, carrying out characteristic extraction on the characteristic patterns; and combining the characteristic features with each other characteristic features to obtain the first characteristic pattern.
read more
Abstract: The embodiment of the invention provides an image processing method and device, a storage medium, a computer program and electronic equipment. The image processing method comprises the following stepsthat: obtaining the characteristic pattern of an image to be detected; through a neural network, on the basis of at least two different scales, carrying out characteristic extraction on the characteristic pattern, and obtaining at least two other characteristic patterns; and combining the characteristic patterns with each other characteristic pattern to obtain the first characteristic pattern ofthe image to be detected. By use of the technical scheme of the embodiment of the invention, the neural network can be used for learning and extracting the characteristics of different scales, and characteristic extraction accuracy and robustness can be improved.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Image processing method and device, computer equipment and storage medium
Sun Shuyang,Jiangmiao Pang,Jianping Shi,Yi Shuai,Wanli Ouyang +4 more
- 30 Nov 2018
TL;DR: In this article, a neural network model is trained through combining features, with a first resolution, output at different segments of the neural network and taking the combined features as inputs of segments with a second resolution corresponding to the different segments; and the processing result of the to-be-processed image is finally output.
7
Patent
Key point detection method and device, electronic equipment and storage medium
Yang Kunlin,Tian Maoqing,Shuai Yi +2 more
- 12 Apr 2019
TL;DR: In this article, the authors proposed a key point detection method based on a pyramid neural network (PHN) for keypoint detection, which comprises the steps: obtaining first feature maps of multiple scales for an input image, and enabling the scales of the first feature map to be in a multiple relation; using a first pyramid neural networks to carry out forward processing on each first feature mapping to obtain second feature maps in one-to-one correspondence with the first map, the second feature mapping being the same as the first mapping in scale.
7
Patent
Image processing method and apparatus, electronic device, storage medium, and program product
Zhao Hengshuang,Yi Zhang,Shi Jianping +2 more
- 13 Feb 2020
TL;DR: In this article, the authors proposed a method for generating a feature map of a to-be-processed image by performing feature extraction on the image, determining a feature weight corresponding to each of a plurality of feature points comprised in the feature map, and obtaining a feature-enhanced feature map by separately transmitting feature information of each feature point to associated other feature points based on the corresponding feature weight.
2
Patent
Network module, distribution method and device, electronic equipment and storage medium
Yi Li,Zhanghui Kuang,Yimin Chen,Zhang Wei +3 more
- 18 Jun 2019
TL;DR: In this paper, a network module consisting of a first network layer, a second network layer and a third network layer which are sequentially cascaded is described. And the purpose of directly constructing a plurality of feature maps with different scales in the network module can be achieved.
1
Patent
Mammary gland image recognition method and device
Wei Zikun,Hua Yiwei,Ding Zezhen +2 more
- 23 Jul 2019
TL;DR: In this paper, the mammary gland image recognition method and device relates to the technical field of machine learning, and the method comprises the steps: obtaining a mammary glands image, inputting the mammaries image into a feature extraction module, and obtaining a feature image of the gland image, wherein the feature image comprises N convolution modules, and each convolution module sequentially comprises a first convolution layer and a second convolutional layer.
1
References
•Posted Content
Multi-Scale Context Aggregation by Dilated Convolutions
Fisher Yu,Vladlen Koltun +1 more
TL;DR: In this article, a new convolutional network module is proposed to aggregate multi-scale contextual information without losing resolution, and the architecture is based on the fact that dilated convolutions support exponential expansion of the receptive field without loss of resolution or coverage.
4K
•Posted Content
Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
Hyeonseob Nam,Bohyung Han +1 more
TL;DR: Zhang et al. as discussed by the authors proposed a novel visual tracking algorithm based on the representations from a discriminatively trained Convolutional Neural Network (CNN), which pretrain a CNN using a large set of videos with tracking ground-truths to obtain a generic target representation.
2.2K
•Posted Content
Multi-Context Attention for Human Pose Estimation
TL;DR: This paper proposes to incorporate convolutional neural networks with a multi-context attention mechanism into an end-to-end framework for human pose estimation and designs novel Hourglass Residual Units (HRUs) to increase the receptive field of the network.
384
Patent
Deep Image-to-Image Network Learning for Medical Image Analysis
S. Kevin Zhou,Dorin Comaniciu,Bogdan Georgescu,Yefeng Zheng,David Liu,Daguang Xu +5 more
- 16 Dec 2016
TL;DR: In this paper, a method and apparatus for automatically performing medical image analysis tasks using deep image-to-image network (DI2IN) learning is presented, where an input medical image of a patient is received.
66
Patent
Face recognition method, device, system and apparatus based on convolutional neural network
Zhu Yue,Jia Jie,Xing Xiaoran +2 more
- 14 Jul 2017
TL;DR: In this article, a face recognition method, device, system and apparatus based on a convolutional neural network (CNN) is described, which comprises the following steps: S1, face detection: using a multilayer CNN feature architecture; S2, key point positioning: obtaining the key point positions of the face by using connecting reference frame regression networks in a cascaded way in deep learning; S3, preprocessing: obtaining a face image with a fixed size; S4, feature extraction: obtaining feature representative vector by means of a feature extraction model; and S
59