Multi-view semi-supervised learning for image classification

Zhu, Shiping; Sun, Xian; Jin, Dongliang

doi:10.1016/j.neucom.2016.02.072

Cited by 40 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the MSRC-v1 dataset, the index number of the randomly extracted images are list = [1,5,12,19,23,28]…”

Section: ) Datasetsmentioning

confidence: 99%

“…Nowadays, multi-view data, which is obtained from different views, has been widely applied in many scientific fields. [1] Among these applications, multi-view semi-supervised classification method has attracted much attention due to its extraordinary performance. Semi-supervised classification is to use a small number of labeled samples [2] and their category information and data distribution information of unlabeled samples to establish a classification model, then classify unknown sample data sets [3].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Semi-Supervised Classification by Joint Global and Local Graph

2019

View full text Add to dashboard Cite

Recently, semisupervised classification methods have been widely applied in many scientific fields, among them, the graph-based methods of semisupervised classification have attracted great attention due to their effectiveness. However, most existing graph-based multi-view methods only construct a graph on each view, and cannot make full use of the abundant information provided by multi-view data. In addition, some semisupervised classification methods directly learn a fixed graph with original input data which contain noise that result in unreliable graphs. Based on these considerations, we propose a novel adaptive multi-view semisupervised classification model which construct a joint global-local graph with total views to attain the connection between each view and within each view. Our model can obtain a better consistency structure and learn the weight coefficient automatically after finite iterations. We also develop a nongreedy optimization algorithm to solve our objective function. In our model, we studied the effect of the global structure on the overall classification performance, and the experiments verified the theoretical derivation. We also studied the effect of label ratio on classification accuracy in different algorithms. The experimental results show that the classification accuracy of our algorithm is higher than other existing algorithms in the range of moderate label ratio so that it has great advantages in practical applications. When the global graph is added, the model still has good performance in terms of running time and convergence rate. The comprehensive experiments on different real-world datasets show the effectiveness of the proposed approach and demonstrate the advantage over other methods.

show abstract

“…On the MSRC-v1 dataset, the index number of the randomly extracted images are list = [1,5,12,19,23,28]…”

Section: ) Datasetsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive Semi-Supervised Classification by Joint Global and Local Graph

2019

View full text Add to dashboard Cite

show abstract

“…During the last decade, several approaches have been proposed to improve the high-level feature extraction in pattern recognition area. Convolutional Neural Network (CNN, or ConvNet) has recently utilized as a powerful class of models for feature extraction purpose such as VGG16, VGG19 and ResNet50 [16], [21]. In this study.…”

Section: ) High-level Featuresmentioning

confidence: 99%

An Effective Automatic Image Annotation Model Via Attention Model and Data Equilibrium

Vatani¹,

Ahvanooey²,

Rahimi³

2018

ijacsa

View full text Add to dashboard Cite

Nowadays, a huge number of images are available. However, retrieving a required image for an ordinary user is a challenging task in computer vision systems. During the past two decades, many types of research have been introduced to improve the performance of the automatic annotation of images, which are traditionally focused on content-based image retrieval. Although, recent research demonstrates that there is a semantic gap between content-based image retrieval and image semantics understandable by humans. As a result, existing research in this area has caused to bridge the semantic gap between low-level image features and high-level semantics. The conventional method of bridging the semantic gap is through the automatic image annotation (AIA) that extracts semantic features using machine learning techniques. In this paper, we propose a novel AIA model based on the deep learning feature extraction method. The proposed model has three phases, including a feature extractor, a tag generator, and an image annotator. First, the proposed model extracts automatically the high and low-level features based on dual tree continues wavelet transform (DT-CWT), singular value decomposition, distribution of color ton, and the deep neural network. Moreover, the tag generator balances the dictionary of the annotated keywords by a new logentropy auto-encoder (LEAE) and then describes these keywords by word embedding. Finally, the annotator works based on the long-short-term memory (LSTM) network in order to obtain the importance degree of specific features of the image. The experiments conducted on two benchmark datasets confirm that the superiority of proposed model compared to the previous models in terms of performance criteria. Keywords-Automatic image annotation; attention model; skewed learning; deep learning, word embedding; log-entropy auto encoderI.

show abstract

“…Since there is no standard human pose depth image library, we builds a data set, including common human actions such as running, jumping, lifting, bending, knee flexion, and interaction. Random forest learning algorithm belongs to supervised learning; the data samples are a known category, and these samples need to be tagged [36][37][38][39]. The tagging method is to divide the human body into 11 parts, and the rest is the background; the approximate position of each part of the human body in the depth image is observed, and then, the position is tagged with the corresponding color.…”

Section: Tagging Body Partsmentioning

confidence: 99%

Human Pose Recognition Based on Depth Image Multifeature Fusion

Wang

Zhou

et al. 2018

Complexity

View full text Add to dashboard Cite

The recognition of human pose based on machine vision usually results in a low recognition rate, low robustness, and low operating efficiency. That is mainly caused by the complexity of the background, as well as the diversity of human pose, occlusion, and selfocclusion. To solve this problem, a feature extraction method combining directional gradient of depth feature (DGoD) and local difference of depth feature (LDoD) is proposed in this paper, which uses a novel strategy that incorporates eight neighborhood points around a pixel for mutual comparison to calculate the difference between the pixels. A new data set is then established to train the random forest classifier, and a random forest two-way voting mechanism is adopted to classify the pixels on different parts of the human body depth image. Finally, the gravity center of each part is calculated and a reasonable point is selected as the joint to extract human skeleton. The experimental results show that the robustness and accuracy are significantly improved, associated with a competitive operating efficiency by evaluating our approach with the proposed data set.

show abstract

Multi-view semi-supervised learning for image classification

Cited by 40 publications

References 20 publications

Adaptive Semi-Supervised Classification by Joint Global and Local Graph

Adaptive Semi-Supervised Classification by Joint Global and Local Graph

An Effective Automatic Image Annotation Model Via Attention Model and Data Equilibrium

Human Pose Recognition Based on Depth Image Multifeature Fusion

Contact Info

Product

Resources

About