Showing papers in "Pattern Recognition in 2020"

TL;DR: A simple yet powerful deep network architecture, U2-Net, for salient object detection (SOD), a two-level nested U-structure that enables us to train a deep network from scratch without using backbones from image classification tasks.

TL;DR: The proposed UWCNN model directly reconstructs the clear latent underwater image, which benefits from the underwater scene prior which can be used to synthesize underwater image training data, and can be easily extended to underwater videos for frame-by-frame enhancement.

TL;DR: A comprehensive survey of algorithms proposed for binary neural networks, mainly categorized into the native solutions directly conducting binarization, and the optimized ones using techniques like minimizing the quantization error, improving the network loss function, and reducing the gradient error are presented.

TL;DR: Li et al. as mentioned in this paper proposed a self-training framework for unsupervised domain adaptive re-ID, which iteratively makes guesses for unlabeled target data based on an encoder and trains the encoder based on the guessed labels.

TL;DR: This survey analyzes the latest state-of-the-art research in HAR in recent years, introduces a classification of HAR methodologies, and shows advantages and weaknesses for methods in each category.

TL;DR: PoseGait exploits human 3D pose estimated from images by Convolutional Neural Network as the input feature for gait recognition and design spatio-temporal features from the3D pose to improve the recognition rate.

TL;DR: This paper proposes to apply Long Short-Term Memory in an end-to-end way to model the pedestrian, seen as a sequence of body parts from head to foot, and develops a novel three-branch framework named Deep-Person, which learns highly discriminative features for person Re-ID.

TL;DR: In this paper, two novel approaches are proposed, which combine Long Short-Term Memory networks and Graph Convolutional Networks to learn long short-term dependencies together with graph structure.

TL;DR: This paper proposes an efficient channel selection layer, namely AutoPruner, to find less important filters automatically in a joint training manner and empirically demonstrates that the gradient information of this channel selectionlayer is also helpful for the whole model training.

TL;DR: A general framework to design VAEs suitable for fitting incomplete heterogenous data, which includes likelihood models for real-valued, positive real valued, interval, categorical, ordinal and count data, and allows accurate estimation of missing data is proposed.

TL;DR: A novel deep multi-view clustering model is proposed by uncovering the hierarchical semantics of the input data in a layer-wise way and utilizing a novel collaborative deep matrix decomposition framework, the hidden representations are learned with respect to different attributes.

TL;DR: The high accuracy obtained with the DSVM validates its efficacy as state-of-the-art algorithm for hyperspectral image classification.

TL;DR: The results show that this ability to simultaneously train models, which exchange knowledge while preserving diversity, leads to state-of-the-art results on two challenging medical image datasets.

TL;DR: A novel hierarchical dense connection network (HDN) is advocated for image SR that outperforms the state-of-the-art methods in terms of quantitative indicators and realistic visual effects, as well as enjoys a fast and accurate reconstruction.

TL;DR: A novel method to implicitly model the relationship among regions of interest in an image with a graph neural network, as well as a novel context-aware attention mechanism to guide attention selection by fully memorizing previously attended visual content are proposed.

TL;DR: This study reveals that deep features provide prominent semantic information and a variety of contextual contents, which contribute to its superior performance in detecting small or occluded objects.

TL;DR: This paper proposes a correlation-aware adversarial DA and DG framework where the features of the source and target data are minimized using correlation alignment along with adversarial learning to achieve a more domain agnostic model.

TL;DR: This paper proposes a spatio-temporal deformable ConvNet module with an attention mechanism, which takes into consideration the mutual correlations in both temporal and spatial domains, to effectively capture the long-range and long-distance dependencies in the video actions.

TL;DR: Tracklet-Plane Matching (TPM), a new approach which improves the performance of MOT by modeling and reducing the interferences from noisy or confusing object detections, outperforms the state-of-the-art MOT methods.

TL;DR: An Identity-Aware CycleGAN (IACycleGAN) model is proposed that applies a new perceptual loss to supervise the image generation network and improves CycleGAN on photo-sketch synthesis by paying more attention to the synthesis of key facial regions, such as eyes and nose, which are important for identity recognition.

TL;DR: A random forest of heterogeneous oblique decision trees that employ several linear classifiers at each non-leaf node on some top ranked partitions which are obtained via one-vs-all and two-hyperclasses based approaches and ranked based on ideal Gini scores and cluster separability is proposed.

TL;DR: A novel image inpainting method for large-scale irregular masks is proposed with a special multistage attention module that considers structure consistency and detail fineness and adopts a partial convolution strategy to avoid the misuse of invalid data during convolution.

TL;DR: In the proposed GLTA framework, the tensor singular value decomposition-based tensor nuclear norm is adopted to explore the high-order cross-view correlations and the manifold regularization is exploited to preserve the local structures embedded in high-dimensional space.

TL;DR: This paper proposes a novel approach for Simultaneous Localization and Mapping by fusing natural and artificial landmarks that integrates both approaches in order to achieve long-term robust tracking in many scenarios.

TL;DR: A reduced universum twin support vector machine for class imbalance learning (RUTSVM-CIL) is proposed in this paper, for the first time, universum learning is incorporated with SVM to solve the problem of class imbalance.

TL;DR: In this article, the authors examined the possibility of utilizing the concept of mutual class potential, used to guide the oversampling process in RBO, in the undersampling procedure.

TL;DR: This paper proposed a fully convolutional network without any recurrent connections trained with the CTC loss function, which achieved state-of-the-art results on seven public benchmark datasets, covering a wide spectrum of text recognition tasks.

TL;DR: The proposed point attention network consists of an encoder and decoder which, together with the LAE-Conv layers and the point-wise spatial attention modules, make it an end-to-end trainable network for predicting dense labels for 3D point cloud segmentation.

TL;DR: In this paper, the authors propose a principled approach for one-class SVMs (OC-SVM) that can be rewritten as distance/pooling neural networks, and apply deep Taylor decomposition (DTD), a methodology that leverages the model structure in order to quickly and reliably explain decisions in terms of input features.

TL;DR: A novel hybrid approach for crack detection in raw images, which combines deep learning models and Bayesian probabilistic analysis for robust crack detection is proposed, which outperforms the state-of-the-art baseline approach on deep CNN classifier.