Patent
Automatic image annotation method integrating depth features and semantic neighborhood
Ke Xiao,Zhou Mingke +1 more
- 21 Dec 2016
7
TL;DR: In this paper, a unified and adaptive depth feature extraction framework based on a depth convolutional neural network (CNN) is built; then, a training set is grouped semantically, and a neighborhood image set of an image to be annotated is built, and finally, the contribution value of each label of the neighborhood images is calculated according to the visual distance.
read more
Abstract: The invention relates to an automatic image annotation method integrating depth features and semantic neighborhood. In view of the problem that manual selection of features takes time and energy in the traditional image annotation method, the problem that the traditional label propagation algorithm ignores semantic neighborhood, which results in visual similarity and semantic dissimilarity and affects the annotation result, and the like, the invention puts forward an automatic image annotation method integrating depth features and semantic neighbors. First, a unified and adaptive depth feature extraction framework based on a depth convolutional neural network (CNN) is built; then, a training set is grouped semantically, and a neighborhood image set of an image to be annotated is built; and finally, the contribution value of each label of the neighborhood images is calculated according to the visual distance, and the contribution values are sorted to get annotation keywords. The method is simple and flexible, and is of strong practicability.
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
Character image identifying and labeling method and system
Weng Yu
- 23 Jan 2018
TL;DR: In this article, a character image identification and labeling method and system is described. But the method is based on the Laplace additional information-against neural network model and is not suitable for image types.
11
Patent
An image automatic marking method and a device based on depth learning
Cheng Qimin,Xu Yuan,Qian Zhang,Kang Shao,Li Sen,Jinling Li +5 more
- 25 Jan 2019
TL;DR: In this article, an image automatic marking method and a device based on depth learning is described. But the method comprises the following steps: extracting visual features of an image to be marked by using depth learning technology; constructing the candidate tag set of the image to annotated by using the image library, and extracting the semantic features of the images to be annotated.
3
Patent
A pedestrian re-identification system and method
Huang Zhiyong,Wang Yujie,Lin Shuang,Yu Zhi,Li Yinsong,Sun Daming +5 more
- 10 May 2019
TL;DR: In this article, a pedestrian re-identification system comprises a multi-stream feature distance fusion system used for calculating the image similarity between an image p to be detected and each reference image in an initial reference image set G, a sorting system for sorting the reference images according to image similarity, and a re-sorting system based on k neighborhood distribution scores.
2
Patent
Image feature extraction method, target recognition method, and corresponding device
Fan Xing,Zhang Xuan,Zhang Chi +2 more
- 25 Dec 2018
TL;DR: In this article, the authors proposed an image feature extraction method, an object recognition method and a corresponding device, which comprises the following steps: extracting features from the first image by using a first convolution neural network to obtain a first feature image; obtaining a plurality of first spatial features based on the first feature images; the second convolutional neural network being used to amplify the number of channels of the first characteristic image to obtain the first amplified characteristic image after the numbers of channels is amplified.
2
Patent
Multimedia resource searching method and device and storage medium
Chen Zhenhong
- 05 Jun 2020
TL;DR: In this article, the authors proposed a method to find a first vector corresponding to a multimedia resource in response to a search request of the multimedia resource, wherein the first vector is used for representing semantic information and visual information in the multimedia resources.
References
Patent
Semantic annotation method for hyperspectral remote sensing image
Jiang Zhiguo,Yang Junli,Zhang Haopeng,Shi Zhenwei +3 more
- 06 Jul 2016
TL;DR: In this article, a semantic annotation method for a hyperspectral remote sensing image is proposed, which comprises the following steps of: I, acquiring training data and test data of the HRS image through spectral information and an annotated truth value of the HRRS image; II, constructing a convolutional neural network according to the number of bands of the HS image; III, training the CNN through the training data to obtain a CNN model; IV, classifying the test data through the CNN model to obtain the semantic annotation result; V, constructing the unary potential-
28
Image annotation based on feature fusion and semantic similarity
Xiaochun Zhang,Chuancai Liu +1 more
TL;DR: Experiments showed that the proposed multi-feature fusion method removed the effects of scale and the correlations of feature distances, so it could represent the total distance better and find the nearest neighbors.
23
Patent
Automatic image annotation method based on deep learning and canonical correlation analysis
Limin Zhang,Kai Liu,Deng Xiangyang,Sun Yongwei,Zhang Jianting +4 more
- 29 Apr 2015
TL;DR: In this article, the authors proposed an automatic image annotation method based on deep learning and canonical correlation analysis, which includes: using a depth Boltzmann machine to extract the high-level feature vectors of images and annotation words, selecting multiple Bernoulli distribution to fit annotation word samples, and selecting Gaussian distributions to fit image features; performing Canonical correlation analysis on the highlevel features of the images and the annotation words; calculating the Mahalanobis distance between to-be-annotated images and training set images in canonical variable space, and performing weighted calculation according to
14
Patent
Image automatic marking method based on Monte Carlo data balance
Ke Xiao,Du Mingzhi,Zhou Mingke +2 more
- 22 Jun 2016
TL;DR: In this article, the authors proposed an image automatic marking method based on Monte Carlo data balance, which comprises the steps of carrying out the region segmentation on the training sample images in a public image library, enabling the segmented regions possessing different characteristic description to correspond to one marking word, and finally inputting the extracted characteristic vectors in a robustness least squares increment limit learning machine.
12
Multi-scale salient region and relevant visual keywords based model for automatic image annotation
Xiao Ke,Wenzhong Guo +1 more
TL;DR: An image annotation model based on multi-scale salient region and relevant visual keywords is proposed that can improve the object descriptions of images and image regions and be used to improve the annotation performance.
5
Related Papers (5)
Liang Sun,Wang Xuelian,Ge Hongwei,Guozhen Tan +3 more
- 27 Jul 2016
Jing Zhang,Weiwei Hu +1 more
- 17 Jan 2013