scholarly journals The Image Annotation Refinement in Embedding Feature Space based on Mutual Information

2022 ◽  
Vol 16 ◽  
pp. 191-201
Author(s):  
Wei Li ◽  
Haiyu Song ◽  
Hongda Zhang ◽  
Houjie Li ◽  
Pengjie Wang
Keyword(s):  
Mutual Information ◽  
Image Annotation ◽  
State Of The Art ◽  
Feature Space ◽  
Semantic Space ◽  
Visual Features ◽  
Novel Approach ◽  
Proposed Model ◽  
Information Method ◽  
Available Information

The ever-increasing size of images has made automatic image annotation one of the most important tasks in the fields of machine learning and computer vision. Despite continuous efforts in inventing new annotation algorithms and new models, results of the state-of-the-art image annotation methods are often unsatisfactory. In this paper, to further improve annotation refinement performance, a novel approach based on weighted mutual information to automatically refine the original annotations of images is proposed. Unlike the traditional refinement model using only visual feature, the proposed model use semantic embedding to properly map labels and visual features to a meaningful semantic space. To accurately measure the relevance between the particular image and its original annotations, the proposed model utilize all available information including image-to-image, label-to-label and image-to-label. Experimental results conducted on three typical datasets show not only the validity of the refinement, but also the superiority of the proposed algorithm over existing ones. The improvement largely benefits from our proposed mutual information method and utilizing all available information.

scholarly journals Convolutional Spatial Attention Model for Reading Comprehension with Multiple-Choice Questions

2019 ◽  
Vol 33 ◽  
pp. 6276-6283 ◽  
Author(s):  
Zhipeng Chen ◽  
Yiming Cui ◽  
Wentao Ma ◽  
Shijin Wang ◽  
Guoping Hu
Keyword(s):  
Reading Comprehension ◽  
Mutual Information ◽  
Spatial Attention ◽  
State Of The Art ◽  
Multiple Choice ◽  
Multiple Choice Questions ◽  
Attention Model ◽  
Novel Approach ◽  
Proposed Model ◽  
Machine Reading

Machine Reading Comprehension (MRC) with multiplechoice questions requires the machine to read given passage and select the correct answer among several candidates. In this paper, we propose a novel approach called Convolutional Spatial Attention (CSA) model which can better handle the MRC with multiple-choice questions. The proposed model could fully extract the mutual information among the passage, question, and the candidates, to form the enriched representations. Furthermore, to merge various attention results, we propose to use convolutional operation to dynamically summarize the attention values within the different size of regions. Experimental results show that the proposed model could give substantial improvements over various state-of- the-art systems on both RACE and SemEval-2018 Task11 datasets.

scholarly journals Zero Shot Learning via Low-rank Embedded Semantic AutoEncoder

2018 ◽  
Author(s):  
Yang Liu ◽  
Quanxue Gao ◽  
Jin Li ◽  
Jungong Han ◽  
Ling Shao
Keyword(s):  
State Of The Art ◽  
Feature Space ◽  
Original Data ◽  
Semantic Space ◽  
Dimensional Subspace ◽  
Low Rank ◽  
Visual Features ◽  
Visual Feature ◽  
Benchmark Datasets ◽  
Low Dimensional

Zero-shot learning (ZSL) has been widely researched and get successful in machine learning. Most existing ZSL methods aim to accurately recognize objects of unseen classes by learning a shared mapping from the feature space to a semantic space. However, such methods did not investigate in-depth whether the mapping can precisely reconstruct the original visual feature. Motivated by the fact that the data have low intrinsic dimensionality e.g. low-dimensional subspace. In this paper, we formulate a novel framework named Low-rank Embedded Semantic AutoEncoder (LESAE) to jointly seek a low-rank mapping to link visual features with their semantic representations. Taking the encoder-decoder paradigm, the encoder part aims to learn a low-rank mapping from the visual feature to the semantic space, while decoder part manages to reconstruct the original data with the learned mapping. In addition, a non-greedy iterative algorithm is adopted to solve our model. Extensive experiments on six benchmark datasets demonstrate its superiority over several state-of-the-art algorithms.

scholarly journals Multimodal Summarization with Guidance of Multimodal Reference

2020 ◽  
Vol 34 (05) ◽  
pp. 9749-9756
Author(s):  
Junnan Zhu ◽  
Yu Zhou ◽  
Jiajun Zhang ◽  
Haoran Li ◽  
Chengqing Zong ◽  
...  
Keyword(s):  
Objective Function ◽  
Evaluation Method ◽  
Reference Data ◽  
State Of The Art ◽  
Semantic Space ◽  
Experimental Results ◽  
Model Output ◽  
Proposed Model ◽  
Evaluation Metric

Multimodal summarization with multimodal output (MSMO) is to generate a multimodal summary for a multimodal news report, which has been proven to effectively improve users' satisfaction. The existing MSMO methods are trained by the target of text modality, leading to the modality-bias problem that ignores the quality of model-selected image during training. To alleviate this problem, we propose a multimodal objective function with the guidance of multimodal reference to use the loss from the summary generation and the image selection. Due to the lack of multimodal reference data, we present two strategies, i.e., ROUGE-ranking and Order-ranking, to construct the multimodal reference by extending the text reference. Meanwhile, to better evaluate multimodal outputs, we propose a novel evaluation metric based on joint multimodal representation, projecting the model output and multimodal reference into a joint semantic space during evaluation. Experimental results have shown that our proposed model achieves the new state-of-the-art on both automatic and manual evaluation metrics. Besides, our proposed evaluation method can effectively improve the correlation with human judgments.

scholarly journals Zero-Shot Learning from Adversarial Feature Residual to Compact Visual Feature

2020 ◽  
Vol 34 (07) ◽  
pp. 11547-11554
Author(s):  
Bo Liu ◽  
Qiulei Dong ◽  
Zhanyi Hu
Keyword(s):  
State Of The Art ◽  
Feature Space ◽  
Visual Features ◽  
Selection Strategy ◽  
Semantic Features ◽  
Visual Feature ◽  
Adversarial Network ◽  
Benchmark Datasets ◽  
Residual Generator ◽  
Object Features

Recently, many zero-shot learning (ZSL) methods focused on learning discriminative object features in an embedding feature space, however, the distributions of the unseen-class features learned by these methods are prone to be partly overlapped, resulting in inaccurate object recognition. Addressing this problem, we propose a novel adversarial network to synthesize compact semantic visual features for ZSL, consisting of a residual generator, a prototype predictor, and a discriminator. The residual generator is to generate the visual feature residual, which is integrated with a visual prototype predicted via the prototype predictor for synthesizing the visual feature. The discriminator is to distinguish the synthetic visual features from the real ones extracted from an existing categorization CNN. Since the generated residuals are generally numerically much smaller than the distances among all the prototypes, the distributions of the unseen-class features synthesized by the proposed network are less overlapped. In addition, considering that the visual features from categorization CNNs are generally inconsistent with their semantic features, a simple feature selection strategy is introduced for extracting more compact semantic visual features. Extensive experimental results on six benchmark datasets demonstrate that our method could achieve a significantly better performance than existing state-of-the-art methods by ∼1.2-13.2% in most cases.

Dual discriminative auto-encoder network for zero shot image recognition

2021 ◽  
pp. 1-12
Author(s):  
Haoyue Bai ◽  
Haofeng Zhang ◽  
Qiong Wang
Keyword(s):  
Image Recognition ◽  
Feature Space ◽  
Semantic Space ◽  
High Dimensional ◽  
Visual Features ◽  
Discriminative Feature ◽  
Latent Space ◽  
Embedding Strategy ◽  
Low Dimensional ◽  
Semantic Attributes

Zero Shot learning (ZSL) aims to use the information of seen classes to recognize unseen classes, which is achieved by transferring knowledge of the seen classes from the semantic embeddings. Since the domains of the seen and unseen classes do not overlap, most ZSL algorithms often suffer from domain shift problem. In this paper, we propose a Dual Discriminative Auto-encoder Network (DDANet), in which visual features and semantic attributes are self-encoded by using the high dimensional latent space instead of the feature space or the low dimensional semantic space. In the embedded latent space, the features are projected to both preserve their original semantic meanings and have discriminative characteristics, which are realized by applying dual semantic auto-encoder and discriminative feature embedding strategy. Moreover, the cross modal reconstruction is applied to obtain interactive information. Extensive experiments are conducted on four popular datasets and the results demonstrate the superiority of this method.

scholarly journals Detecting Human-Object Interactions via Functional Generalization

2020 ◽  
Vol 34 (07) ◽  
pp. 10460-10469 ◽  
Author(s):  
Ankan Bansal ◽  
Sai Saketh Rambhatla ◽  
Abhinav Shrivastava ◽  
Rama Chellappa
Keyword(s):  
Experimental Validation ◽  
State Of The Art ◽  
Visual Features ◽  
Average Precision ◽  
Proposed Model ◽  
Human Object ◽  
Significant Performance ◽  
Unseen Objects ◽  
Performance Gains ◽  
Object Interactions

We present an approach for detecting human-object interactions (HOIs) in images, based on the idea that humans interact with functionally similar objects in a similar manner. The proposed model is simple and efficiently uses the data, visual features of the human, relative spatial orientation of the human and the object, and the knowledge that functionally similar objects take part in similar interactions with humans. We provide extensive experimental validation for our approach and demonstrate state-of-the-art results for HOI detection. On the HICO-Det dataset our method achieves a gain of over 2.5% absolute points in mean average precision (mAP) over state-of-the-art. We also show that our approach leads to significant performance gains for zero-shot HOI detection in the seen object setting. We further demonstrate that using a generic object detector, our model can generalize to interactions involving previously unseen objects.

scholarly journals A Variational Autoencoder with Deep Embedding Model for Generalized Zero-Shot Learning

2020 ◽  
Vol 34 (07) ◽  
pp. 11733-11740
Author(s):  
Peirong Ma ◽  
Xiao Hu
Keyword(s):  
Feature Space ◽  
Semantic Space ◽  
Image Features ◽  
Visual Feature ◽  
Softmax Classifier ◽  
Proposed Model ◽  
Variational Autoencoder ◽  
Deep Embedding ◽  
Latent Features ◽  
Generalized Zero

Generalized zero-shot learning (GZSL) is a challenging task that aims to recognize not only unseen classes unavailable during training, but also seen classes used at training stage. It is achieved by transferring knowledge from seen classes to unseen classes via a shared semantic space (e.g. attribute space). Most existing GZSL methods usually learn a cross-modal mapping between the visual feature space and the semantic space. However, the mapping model learned only from the seen classes will produce an inherent bias when used in the unseen classes. In order to tackle such a problem, this paper integrates a deep embedding network (DE) and a modified variational autoencoder (VAE) into a novel model (DE-VAE) to learn a latent space shared by both image features and class embeddings. Specifically, the proposed model firstly employs DE to learn the mapping from the semantic space to the visual feature space, and then utilizes VAE to transform both original visual features and the features obtained by the mapping into latent features. Finally, the latent features are used to train a softmax classifier. Extensive experiments on four GZSL benchmark datasets show that the proposed model significantly outperforms the state of the arts.

scholarly journals POS-Taggging Malay Corpus: A Novel Approach Based on Maximum Entropy

2018 ◽  
Vol 7 (3.20) ◽  
pp. 6
Author(s):  
Juhaida Abu Bakar ◽  
Khairuddin Khairuddin ◽  
Mohammad Faidzul Nasrudin ◽  
Mohd Zamri Murah
Keyword(s):  
Maximum Entropy ◽  
Text Analysis ◽  
State Of The Art ◽  
The State ◽  
The Past ◽  
Novel Approach ◽  
Proposed Model ◽  
Automated Text Analysis ◽  
Development Of Language ◽  
Language Technologies

Jawi and Roman scripts are represented Malay language. In the past, Jawi writings are widely used by the Malay community and foreigners; and it can be seen in the old documents. Old documents face the risk of background damage. In order to preserve this valuable information, there are significant needs to automated Jawi materials. Based on previous literature, POS-tags are known as the first phase in the automated text analysis; and the development of language technologies can barely initiate without this phase. We highlight the existing POS-tags approaches; and suggest the development of Malay Jawi POS-tags using extended ME-based approach on NUWT Corpus. Results have shown that the proposed model yielded a higher accuracy in comparison to the state-of-the-art model.  

scholarly journals Generative Image Inpainting with Submanifold Alignment

2019 ◽  
Author(s):  
Ang Li ◽  
Jianzhong Qi ◽  
Rui Zhang ◽  
Xingjun Ma ◽  
Kotagiri Ramamohanarao
Keyword(s):  
State Of The Art ◽  
Image Inpainting ◽  
Feature Space ◽  
Original Data ◽  
Local Data ◽  
Deep Feature ◽  
Proposed Model ◽  
Benchmark Datasets ◽  
Structural Consistency ◽  
Different Levels

Image inpainting aims at restoring missing regions of corrupted images, which has many applications such as image restoration and object removal. However, current GAN-based generative inpainting models do not explicitly exploit the structural or textural consistency between restored contents and their surrounding contexts. To address this limitation, we propose to enforce the alignment (or closeness) between the local data submanifolds (subspaces) around restored images and those around the original (uncorrupted) images during the learning process of GAN-based inpainting models. We exploit Local Intrinsic Dimensionality (LID) to measure, in deep feature space, the alignment between data submanifolds learned by a GAN model and those of the original data, from a perspective of both images (denoted as iLID) and local patches (denoted as pLID) of images. We then apply iLID and pLID as regularizations for GAN-based inpainting models to encourage two different levels of submanifold alignments: 1) an image-level alignment to improve structural consistency, and 2) a patch-level alignment to improve textural details. Experimental results on four benchmark datasets show that our proposed model can generate more accurate results than state-of-the-art models.

scholarly journals MULTI-FEATURE MUTUAL INFORMATION IMAGE REGISTRATION

2012 ◽  
Vol 31 (1) ◽  
pp. 43 ◽  
Author(s):  
Dejan Tomaževič ◽  
Boštjan Likar ◽  
Franjo Pernuš
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Spatial Information ◽  
Similarity Measures ◽  
Feature Space ◽  
Efficient Estimation ◽  
Mr Images ◽  
Feature Correspondence ◽  
Information Theoretic ◽  
Novel Approach

Nowadays, information-theoretic similarity measures, especially the mutual information and its derivatives, are one of the most frequently used measures of global intensity feature correspondence in image registration. Because the traditional mutual information similarity measure ignores the dependency of intensity values of neighboring image elements, registration based on mutual information is not robust in cases of low global intensity correspondence. Robustness can be improved by adding spatial information in the form of local intensity changes to the global intensity correspondence. This paper presents a novel method, by which intensities, together with spatial information, i.e., relations between neighboring image elements in the form of intensity gradients, are included in information-theoretic similarity measures. In contrast to a number of heuristic methods that include additional features into the generic mutual information measure, the proposed method strictly follows information theory under certain assumptions on feature probability distribution. The novel approach solves the problem of efficient estimation of multifeature mutual information from sparse high-dimensional feature space. The proposed measure was tested on magnetic resonance (MR) and computed tomography (CT) images. In addition, the measure was tested on positron emission tomography (PET) and MR images from the widely used Retrospective Image Registration Evaluation project image database. The results indicate that multi-feature mutual information, which combines image intensities and intensity gradients, is more robust than the standard single-feature intensity based mutual information, especially in cases of low global intensity correspondences, such as in PET/MR images or significant intensity inhomogeneity.

Sign in / Sign up

Export Citation Format

Share Document