- Research Article
- 10.4225/03/58b60c41254dc
Learning discriminative relational features for sequence labeling
- Feb 28, 2017
- arXiv: Learning
- Naveen Nair
Learning discriminative relational features for sequence labeling
This paper addresses the problem of recognizing human actions from RGB-D videos. A discriminative relational feature learning method is proposed for fusing heterogeneous RGB and depth modalities, and classifying the actions in RGB-D sequences. Our method factorizes the feature matrix of each modality, and enforces the same semantics for them in order to learn shared features from multimodal data. This allows us to capture the complex correlations between the two modalities. To improve the discriminative power of the relational features, we introduce a hinge loss to measure the classification accuracy when the features are employed for classification. This essentially performs supervised factorization, and learns discriminative features that are optimized for classification. We formulate the recognition task within a maximum margin framework, and solve the formulation using a coordinate descent algorithm. The proposed method is extensively evaluated on two public RGB-D action data sets. We demonstrate that the proposed method can learn extremely low-dimensional features with superior discriminative power, and outperforms the state-of-the-art methods. It also achieves high performance when one modality is missing in testing or training.
Learning discriminative relational features for sequence labeling
Learning discriminative relational features for sequence labeling
Learning Discriminative Features with Region Attention and Refinement Network for Facial Expression Recognition in the Wild
Facial Expression Recognition (FER) has achieved significant performance in recent years. However, learning discriminative expression features is still a pending issue, due to disturbances in the wild and high similarities across different expressions. To remedy this issue, we present an innovative region attention and refinement network (RARN) capable of learning discriminative and robust features for accurate FER in the wild. Specifically, RARN mainly learn discriminative features from two different aspects: Multi-head Local Attention Network (MLAN) and a Latent Feature Mining Network (LFMN). MLAN can accurately locate the regions of interest and extract robust features in the wild by employing two modules, in which the region generation module divides the midlevel feature maps into several subregions and local attention module highlights the local discriminative features of regions of interest through attention block. LFMN combines a series of latent expression features associated with their corresponding importance weights to further obtain the final discriminative expression features. Experimental results on three widely-used in-the-wild FER datasets demonstrate proposed algorithm achieves favorable performance against state-of-the-art methods. The source code will be found at https://github.com/lizoe/RARN.
Read moreJoint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation
Recently, considerable effort has been devoted to deep domain adaptation in computer vision and machine learning communities. However, most of existing work only concentrates on learning shared feature representation by minimizing the distribution discrepancy across different domains. Due to the fact that all the domain alignment approaches can only reduce, but not remove the domain shift, target domain samples distributed near the edge of the clusters, or far from their corresponding class centers are easily to be misclassified by the hyperplane learned from the source domain. To alleviate this issue, we propose to joint domain alignment and discriminative feature learning, which could benefit both domain alignment and final classification. Specifically, an instance-based discriminative feature learning method and a center-based discriminative feature learning method are proposed, both of which guarantee the domain invariant features with better intra-class compactness and inter-class separability. Extensive experiments show that learning the discriminative features in the shared feature space can significantly boost the performance of deep domain adaptation methods.
Read moreFace Alignment by Discriminative Feature Learning
In this paper, we study the effect of discriminative feature learning for face alignment. We claim that features of the same facial landmarks on different images should share similarities at the feature-level. Thus, we propose the Discriminative Feature Learning method (DFL) for face alignment based on the Fully Convolutional Network (FCN). First, the face image is aligned frontal to eliminate the effect of pose and scale. Second, landmark-specific features are extracted from feature maps of the FCN. A distance constraint on landmark features is added to learn discriminative landmark features. Our experiment results show that DFL can effectively improve the performance of face alignment.
Read moreImproving dense conditional random field for retinal vessel segmentation by discriminative feature learning and thin-vessel enhancement
Improving dense conditional random field for retinal vessel segmentation by discriminative feature learning and thin-vessel enhancement
Read moreSpectral–Spatial Discriminant Feature Learning for Hyperspectral Image Classification
Sparse representation classification (SRC) is being widely applied to target detection in hyperspectral images (HSI). However, due to the problem in HSI that high-dimensional data contain redundant information, SRC methods may fail to achieve high classification performance, even with a large number of spectral bands. Selecting a subset of predictive features in a high-dimensional space is an important and challenging problem for hyperspectral image classification. In this paper, we propose a novel discriminant feature learning (DFL) method, which combines spectral and spatial information into a hypergraph Laplacian. First, a subset of discriminative features is selected, which preserve the spectral structure of data and the inter- and intra-class constraints on labeled training samples. A feature evaluator is obtained by semi-supervised learning with the hypergraph Laplacian. Secondly, the selected features are mapped into a further lower-dimensional eigenspace through a generalized eigendecomposition of the Laplacian matrix. The finally extracted discriminative features are used in a joint sparsity-model algorithm. Experiments conducted with benchmark data sets and different experimental settings show that our proposed method increases classification accuracy and outperforms the state-of-the-art HSI classification methods.
Read moreDiscriminative supplementary representation learning for novel-category classification
Discriminative supplementary representation learning for novel-category classification
Learning informative and discriminative semantic features for robust facial expression recognition
Learning informative and discriminative semantic features for robust facial expression recognition
Max-Margin-Based Discriminative Feature Learning.
In this brief, we propose a new max-margin-based discriminative feature learning method. In particular, we aim at learning a low-dimensional feature representation, so as to maximize the global margin of the data and make the samples from the same class as close as possible. In order to enhance the robustness to noise, we leverage a regularization term to make the transformation matrix sparse in rows. In addition, we further learn and leverage the correlations among multiple categories for assisting in learning discriminative features. The experimental results demonstrate the power of the proposed method against the related state-of-the-art methods.
Read moreDiscriminative Learning for Alzheimer's Disease Diagnosis via Canonical Correlation Analysis and Multimodal Fusion
To address the challenging task of diagnosing neurodegenerative brain disease, such as Alzheimer's disease (AD) and mild cognitive impairment (MCI), we propose a novel method using discriminative feature learning and canonical correlation analysis (CCA) in this paper. Specifically, multimodal features and their CCA projections are concatenated together to represent each subject, and hence both individual and shared information of AD disease are captured. A discriminative learning with multilayer feature hierarchy is designed to further improve performance. Also, hybrid representation is proposed to maximally explore data from multiple modalities. A novel normalization method is devised to tackle the intra- and inter-subject variations from the multimodal data. Based on our extensive experiments, our method achieves an accuracy of 96.93% [AD vs. normal control (NC)], 86.57 % (MCI vs. NC), and 82.75% [MCI converter (MCI-C) vs. MCI non-converter (MCI-NC)], respectively, which outperforms the state-of-the-art methods in the literature.
Read moreDiscriminative feature learning for underwater fish recognition
Underwater fish recognition is an important task in fish stock assessment and marine ecosystem studies. Machine learning techniques have been applied to train high-performance fish recognition models from underwater images. However, underwater images often contain extremely noisy backgrounds, hindering the training of accurate recognition models. Traditional methods exploit handcrafted features to train traditional classifiers. These methods often suffer from low recognition accuracy and limited scalability to large-scale datasets. While deep learning approaches have been proposed, the challenge of learning with noisy underwater images has not yet been fully addressed. We propose a discriminative feature learning (DFL) framework to train accurate fish recognition models on noisy underwater images. By leveraging the idea of contrastive learning, DFL encourages the model to learn more discriminative features for images in different classes and similar features for images in the same class. To better address the noisy background problem, DFL also utilizes a regularization technique called attention suppression to prevent the model from paying too much attention to the noisy background. Experimental results on three benchmark datasets validate the superior performance of DFL over the current state-of-the-art deep learning approaches.
Read moreA Classification Algorithm Based on Discriminative Transfer Feature Learning for Early Diagnosis of Alzheimer’s Disease
This paper proposes a discriminative transfer feature learning method for MCI conversion prediction using data from the target domain and the auxiliary domain. A transfer component analysis method based on the Maximum Mean Discrepancy (MMD) is proposed at first, which is used to weaken the difference of data distribution between the relevant domain and the target domain. Next, the discriminant optimization term is added to measure the correlation between the sample categories and the sample features of the auxiliary domain, and to improve the inter-class separability of the algorithm. Finally, the support vector machine (SVM) is used to classify MCI patients.KeywordsADMild cognitive impairmentTransfer learningSVM
Read moreMulti-Task and Multi-Modal Learning for RGB Dynamic Gesture Recognition
Gesture recognition is getting more and more popular due to various application possibilities in human-machine interaction. Existing multi-modal gesture recognition systems take multi-modal data as input to improve accuracy, but such methods require more modality sensors, which will greatly limit their application scenarios. Therefore we propose an end-to-end multi-task learning framework in training 2D convolutional neural networks. The framework can use the depth modality to improve accuracy during training and save costs by using only RGB modality during inference. Our framework is trained to learn a representation for multi-task learning: gesture segmentation and gesture recognition. Depth modality contains the prior information for the location of the gesture. Therefore it can be used as the supervision for gesture segmentation. A plug-and-play module named Multi-Scale-Decoder is designed to realize gesture segmentation, which contains two sub-decoder. It is used in the lower stage and higher stage respectively, and can help the network pay attention to key target areas, ignore irrelevant information, and extract more discriminant features. Additionally, the MSD module and depth modality are only used in the training stage to improve gesture recognition performance. Only RGB modality and network without MSD are required during inference. Experimental results on three public gesture recognition datasets show that our proposed method provides superior performance compared with existing gesture recognition frameworks. Moreover, using the proposed plug-and-play MSD in other 2D CNN-based frameworks also get an excellent accuracy improvement.
Read moreWorst-Case Discriminative Feature Learning via Max-Min Ratio Analysis.
We propose a novel discriminative feature learning method via Max-Min Ratio Analysis (MMRA) for exclusively dealing with the long-standing "worst-case class separation" problem. Existing technologies simply consider maximizing the minimal pairwise distance on all class pairs in the low-dimensional subspace, which is unable to separate overlapped classes entirely especially when the distribution of samples within same class is diverging. We propose a new criterion, i.e., Max-Min Ratio Analysis (MMRA) that focuses on maximizing the minimal ratio value of between-class and within-class scatter to extremely enlarge the separability on the overlapped pairwise classes. Furthermore, we develop two novel discriminative feature learning models for dimensionality reduction and metric learning based on our MMRA criterion. However, solving such a non-smooth non-convex max-min ratio problem is challenging. As an important theoretical contribution in this paper, we systematically derive an alternative iterative algorithm based on a general max-min ratio optimization framework to solve a general max-min ratio problem with rigorous proofs of convergence. More importantly, we also present another solver based on bisection search strategy to solve the SDP problem efficiently. To evaluate the effectiveness of proposed methods, we conduct extensive pattern classification and image retrieval experiments on several artificial datasets and real-world ScRNA-seq datasets, and experimental results demonstrate the effectiveness of proposed methods.
Read morePiece-wise max-margin-based discriminative feature learning
Max-margin-based discriminative feature learning method (MMLDF) learns a low-dimensional feature representation such that the global margin of the data is maximised and samples of the same class become as close as possible. Non-linear version of MMLDF uses kernel method. In this paper, we first prove that the kernel MMLDF cannot be used for discriminative feature learning. Strictly speaking, the kernel MMLDF maps input vectors into zero vectors which cannot then be classified. Therefore, to overcome this problem, we propose a novel non-linear MMLDF which uses piece-wise learning technique. Then, by using alternating optimisation, an algorithm is proposed to solve our proposed model. Experimental results on real datasets confirm that our proposed piece-wise MMLDF outperforms linear MMLDF and piece-wise SVM.
Read more