Patent
Model training method and device
Wan Shaohua,Yang Song,Chen Zhijun +2 more
- 15 Mar 2017
11
TL;DR: In this paper, a model training method and device for image processing is described, which consists of selecting a training sample set from a test sample set, training to obtain an identification model according to samples included in the training set, the training sample sets being a subset of the test set, including positive samples and negative samples, the number of the negative samples in the trained sample set being smaller than that of the positive ones in the tested sample set.
read more
Abstract: The invention discloses a model training method and device, and belongs to the field of image processing. The method includes selecting a training sample set from a test sample set, training to obtain an identification model according to samples included in the training sample set, the training sample set being a subset of the test sample set, the training sample set including positive samples and negative samples, the number of the negative samples in the training sample set being smaller than that of the negative samples in the test sample set, identifying samples included in the test sample set according to the identification model, and updating the identification model according to samples that are wrongly identified; the problem that since the samples are limited in number, a face identification model obtained by training is not accurate enough is solved; after the identification model is obtained by training a limited number of samples, the identification model is updated again according to the samples that are wrongly identified, the identification model can be constantly optimized, and thus the accuracy is improved when the identification model is used for identification.
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
Method and device of face detection
Sun Xudong,Wu Pengcheng,Xu Zhuhong +2 more
- 12 Dec 2017
TL;DR: In this article, a multi-layer convolutional approach is used to obtain a final external processed image, which contains image detail information after intermediate convolution processing, multi-scale feature fusion is formed, description for the original target image is more accurate, and thus face recognition and position determination are also more accurate.
7
Patent
Terminal feature recognition method and system
Wang Yuedong,Zhang Cong,Zhang Jushi,Mao Dongtao +3 more
- 24 Oct 2017
TL;DR: In this article, the authors presented a terminal feature recognition method and system, which comprises the steps that fingerprint information corresponding to a to-be-recognized terminal in a preset time period is acquired, and the fingerprint information is used as a to be-recognised sample.
5
Patent
Neural network training method and device thereof
Wang Naiyan,Huang Xiukun +1 more
- 19 Dec 2017
TL;DR: In this paper, a neural network training method for improving robustness and accuracy of neural networks is described, which comprises the steps of training the primary neural network model by means of a training image set.
5
Patent
Method, device and storage medium for training classifier based on sample characteristics
Wei Jinxia,Long Chun,Zhao Jing,Yang Fan +3 more
- 15 Jan 2019
TL;DR: In this article, a method, a device and a storage medium for training a classifier based on sample characteristics, which relate to the field of network security, is presented. But the method of the invention comprises the following steps: acquiring a sample data set for training the classifier; selecting N sample data in the sample data sets as a target sample data subset; wherein N is a positive integer smaller than M, and M is the total number of sample data used in the data set; selecting the characteristics of each sample data, and carrying out feature weighting transformation on the
3
Patent
Method and device for training convolutional neural network model
Wan Shaohua
- 22 Sep 2017
TL;DR: In this paper, a method for training a convolutional neural network model for image identification is presented. But the method is limited to image classification and it is not suitable for image classification in general.
2
References
Patent
Multiclass image classification method based on active learning and semi-supervised learning
Yongfeng Cao,Rong Chen,Hui Yin +2 more
- 06 Oct 2010
TL;DR: In this paper, the authors proposed a multiclass image classification method based on active learning and semi-supervised learning, which consists of five steps: initial sample selection and classifier model training, BvSB active learning sample selection, CST semi supervised learning, training sample set and classification model updating and assorting process iteration.
57
Patent
Video feed target tracking
Gerard Medioni,Qian Yu,Thang Ba Dinh +2 more
- 01 Apr 2009
TL;DR: In this article, a generative tracker and a discriminative tracker are used to track an object with information based on an output from the generator and an output of the discriminator.
28
Patent
Classifier training method and device
Ben Guosheng,Li Yan +1 more
- 10 Feb 2016
TL;DR: In this article, a classifier training method and device is described, which comprises the following steps: screening samples from an original type sample set to form a training type sample sets, carrying out training on samples in the training type samples set to obtain a parameter file for prediction, and carrying out prediction on the samples in original type samples according to the parameter file to get a type sample predication result, and adding the samples of which the predication results are wrong, to the training sample set, updating the training test set, retraining the sample in the updated test set to
21
Patent
Training method and device of face recognition model
Zhang Tao,Chen Zhijun,Long Fei +2 more
- 23 Mar 2016
TL;DR: In this paper, the authors reveal a training method and device of a face recognition model, which comprises the following steps: obtaining an original face classification model, wherein the original face recognition models comprises N layers of characteristic coefficients; obtaining a second training sample set, and determining a first original classification label of each sample face image in the second training set; adopting the front M layers of feature vectors of the original feature vectors to code each sampled face image.
14
Patent
Speech recognition model training method, speech recognition model training device and terminal
Guo Meng,Yang Lei,Zhang Jian,Yu Rongrong +3 more
- 26 Mar 2014
TL;DR: In this article, a speech recognition model training method is described, which comprises the steps of acquiring speech of a reading sample of a user to obtain a user speech sample, extracting features of the acquired user speech samples and creating a speech model according to the extracted features.
13
Related Papers (5)
Jiang Xiao-Yan,Cai Ning,Zhang Kai,Chu Wei +3 more
- 05 Sep 2017
Jiang Xiao-Yan,Xu Yang,Dai Bin,Chu Wei +3 more
- 31 Aug 2017
Jiang Xiao-Yan,Xu Yang,Dai Bin,Chu Wei +3 more
- 20 Dec 2018
Yang Shaoxiong,Zhao Chen +1 more
- 11 Jan 2019