Search

Convolutional Neural Networks (CNN) and Graph Neural Networks (GNN), such as Graph Attention Networks (GAT), are two classic neural network models, which are applied to the processing of grid data and graph data respectively. They have achieved outstanding performance in hyperspectral images (HSIs) classification field, which have attracted great interest. However, CNN has been facing the problem of small samples and GNN has to pay a huge computational cost, which restrict the performance of the two models. In this paper, we propose Weighted Feature Fusion of Convolutional Neural Network and Graph Attention Network (WFCG) for HSI classification, by using the characteristics of superpixel-based GAT and pixel-based CNN, which proved to be complementary. We first establish GAT with the help of superpixel-based encoder and decoder modules. Then we combined the attention mechanism to construct CNN. Finally, the features are weighted fusion with the characteristics of two neural network models. Rigorous experiments on three real-world HSI data sets show WFCG can fully explore the high-dimensional feature of HSI, and obtain competitive results compared to other state-of-the art methods. [ABSTRACT FROM AUTHOR]

Articulated hand pose estimation is one of core technologies in human–computer interaction. Despite the recent progress, most existing methods still cannot achieve satisfactory performance, partly due to the difficulty of the embedded high-dimensional nonlinear regression problem. Most existing data-driven methods directly regress 3D hand pose from 2D depth image, which cannot fully utilize the depth information. In this paper, we propose a novel multi-view convolutional neural network (CNN)-based approach for 3D hand pose estimation. To better exploit 3D information in the depth image, we project the point cloud generated from the query depth image onto multiple views of two projection settings and integrate them for more robust estimation. Multi-view CNNs are trained to learn the mapping from projected images to heat-maps, which reflect probability distributions of joints on each view. These multi-view heat-maps are then fused to estimate the optimal 3D hand pose with learned pose priors, and the unreliable information in multi-view heat-maps is suppressed using a view selection method. Experimental results show that the proposed method is superior to the state-of-the-art methods on two challenging data sets. Furthermore, a cross-data set experiment also validates that our proposed approach has good generalization ability. [ABSTRACT FROM AUTHOR]

Sparse representation based classification (SRC), nuclear-norm matrix regression (NMR), and deep learning (DL) have achieved a great success in face recognition (FR). However, there still exist some intrinsic limitations among them. SRC and NMR based coding methods belong to one-step model, such that the latent discriminative information of the coding error vector cannot be fully exploited. DL, as a multi-step model, can learn powerful representation, but relies on large-scale data and computation resources for numerous parameters training with complicated back-propagation. Straightforward training of deep neural networks from scratch on small-scale data is almost infeasible. Therefore, in order to develop efficient algorithms that are specifically adapted for small-scale data, we propose to derive the deep models of SRC and NMR. Specifically, in this paper, we propose an end-to-end deep cascade model (DCM) based on SRC and NMR with hierarchical learning, nonlinear transformation and multi-layer structure for corrupted face recognition. The contributions include four aspects. First, an end-to-end deep cascade model for small-scale data without back-propagation is proposed. Second, a multi-level pyramid structure is integrated for local feature representation. Third, for introducing nonlinear transformation in layer-wise learning, softmax vector coding of the errors with class discrimination is proposed. Fourth, the existing representation methods can be easily integrated into our DCM framework. Experiments on a number of small-scale benchmark FR datasets demonstrate the superiority of the proposed model over state-of-the-art counterparts. Additionally, a perspective that deep-layered learning does not have to be convolutional neural network with back-propagation optimization is consolidated. The demo code is available in https://github.com/liuji93/DCM [ABSTRACT FROM AUTHOR]

In this paper, we focus on X-ray images (X-radiographs) of paintings with concealed sub-surface designs (e.g., deriving from reuse of the painting support or revision of a composition by the artist), which therefore include contributions from both the surface painting and the concealed features. In particular, we propose a self-supervised deep learning-based image separation approach that can be applied to the X-ray images from such paintings to separate them into two hypothetical X-ray images. One of these reconstructed images is related to the X-ray image of the concealed painting, while the second one contains only information related to the X-ray image of the visible painting. The proposed separation network consists of two components: the analysis and the synthesis sub-networks. The analysis sub-network is based on learned coupled iterative shrinkage thresholding algorithms (LCISTA) designed using algorithm unrolling techniques, and the synthesis sub-network consists of several linear mappings. The learning algorithm operates in a totally self-supervised fashion without requiring a sample set that contains both the mixed X-ray images and the separated ones. The proposed method is demonstrated on a real painting with concealed content, Do na Isabel de Porcel by Francisco de Goya, to show its effectiveness. [ABSTRACT FROM AUTHOR]

In recent years, image denoising has benefited a lot from deep neural networks. However, these models need large amounts of noisy-clean image pairs for supervision. Although there have been attempts in training denoising networks with only noisy images, existing self-supervised algorithms suffer from inefficient network training, heavy computational burden, or dependence on noise modeling. In this paper, we proposed a self-supervised framework named Neighbor2Neighbor for deep image denoising. We develop a theoretical motivation and prove that by designing specific samplers for training image pairs generation from only noisy images, we can train a self-supervised denoising network similar to the network trained with clean images supervision. Besides, we propose a regularizer in the perspective of optimization to narrow the optimization gap between the self-supervised denoiser and the supervised denoiser. We present a very simple yet effective self-supervised training scheme based on the theoretical understandings: training image pairs are generated by random neighbor sub-samplers, and denoising networks are trained with a regularized loss. Moreover, we propose a training strategy named BayerEnsemble to adapt the Neighbor2Neighbor framework in raw image denoising. The proposed Neighbor2Neighbor framework can enjoy the progress of state-of-the-art supervised denoising networks in network architecture design. It also avoids heavy dependence on the assumption of the noise distribution. We evaluate the Neighbor2Neighbor framework through extensive experiments, including synthetic experiments with different noise distributions and real-world experiments under various scenarios. The code is available online: https://github.com/TaoHuang2018/Neighbor2Neighbor. [ABSTRACT FROM AUTHOR]

Image super-resolution (SR) task aims to recover high-resolution (HR) images from degraded low-resolution (LR) images, which has achieved great progress due to the recent advances of deep neural networks. Due to severe information loss of the LR images, it is more challenging to reconstruct high quality HR images at large scale factors, i. e., higher than $4\times $. Traditional reference image based SR methods usually perform patch matching to locate detailed texture from HR reference images which could provide fine details from similar image contents. But it suffers from difficulties in achieving good matching in the largely downscaled image space or feature space due to the ill-posed nature between LR and HR mapping. In this paper, we tackle this problem by exploiting fine details contained in reference HR images. Inspired by vector quantization (VQ), we propose a simple yet effective auto-encoder convolutional neural network (CNN) module to learn discrete representations of images. Furthermore, we propose to progressively learn pairs of cross-scale discrete feature representations using paired LR and HR reference images. The coarser scale of the discrete representation is responsible for encoding the global image structure while the paired finer scale of the discrete representation takes charge of capturing missing details in the finer image scale. During inference, continuous features of the test LR image are used as queries to retrieve finer scale discrete representations (value) by searching the nearest coarser scale discrete representations (key). Then, the queries and retrieved values are combined to progressively recover the HR image. Experimental results indicate that when compared with the state-of-the-art image SR models, the proposed method can achieve advanced performance in terms of both objective quality and subjective quality. The code will be available on URL: https://github.com/sunwj/refsr. [ABSTRACT FROM AUTHOR]

Research in light field reconstruction focuses on synthesizing novel views with the assistance of depth information. In this paper, we present a learning-based light field reconstruction approach by fusing a set of sheared epipolar plane images (EPIs). We start by showing that a patch in a sheared EPI will exhibit a clear structure when the sheared value equals the depth of that patch. By taking advantage of this pattern, a convolutional neural network (CNN) is then trained to evaluate the sheared EPIs, and output a reference score for fusing the sheared EPIs. The proposed CNN is elaborately designed to learn the similarity degree between the input sheared EPI and the ground truth EPI. Therefore, no depth information is required for network training and reasoning. We demonstrate the high performance of the proposed method through evaluations on synthetic scenes, real-world scenes, and challenging microscope light fields. We also show a further application of our proposed network for depth inference. [ABSTRACT FROM AUTHOR]

While action recognition has become an important line of research in computer vision, the recognition of particular events, such as aggressive behaviors, or fights, has been relatively less studied. These tasks may be extremely useful in several video surveillance scenarios, such as psychiatric wards, prisons, or even in personal camera smartphones. Their potential usability has led to a surge of interest in developing fight or violence detectors. One of the key aspects in this case is efficiency, that is, these methods should be computationally fast. “Handcrafted” spatio-temporal features that account for both motion and appearance information can achieve high accuracy rates, albeit the computational cost of extracting some of those features is still prohibitive for practical applications. The deep learning paradigm has been recently applied for the first time to this task too, in the form of a 3D convolutional neural network that processes the whole video sequence as input. However, results in human perception of other’s actions suggest that, in this specific task, motion features are crucial. This means that using the whole video as input may add both redundancy and noise in the learning process. In this paper, we propose a hybrid “handcrafted/learned” feature framework which provides better accuracy than the previous feature learning method, with similar computational efficiency. The proposed method is compared to three related benchmark data sets. The method outperforms the different state-of-the-art methods in two of the three considered benchmark data sets. [ABSTRACT FROM AUTHOR]

In steerable filters, a filter of arbitrary orientation can be generated by a linear combination of a set of “basis filters.” Steerable properties dominate the design of the traditional filters, e.g., Gabor filters and endow features the capability of handling spatial transformations. However, such properties have not yet been well explored in the deep convolutional neural networks (DCNNs). In this paper, we develop a new deep model, namely, Gabor convolutional networks (GCNs or Gabor CNNs), with Gabor filters incorporated into DCNNs such that the robustness of learned features against the orientation and scale changes can be reinforced. By manipulating the basic element of DCNNs, i.e., the convolution operator, based on Gabor filters, GCNs can be easily implemented and are readily compatible with any popular deep learning architecture. We carry out extensive experiments to demonstrate the promising performance of our GCNs framework, and the results show its superiority in recognizing objects, especially when the scale and rotation changes take place frequently. Moreover, the proposed GCNs have much fewer network parameters to be learned and can effectively reduce the training complexity of the network, leading to a more compact deep learning model while still maintaining a high feature representation capacity. The source code can be found at https://github.com/bczhangbczhang. [ABSTRACT FROM AUTHOR]

Image restoration methods based on convolutional neural networks have shown great success in the literature. However, since most of networks are not deep enough, there is still some room for the performance improvement. On the other hand, though some models are deep and introduce shortcuts for easy training, they ignore the importance of location and scaling of different inputs within the shortcuts. As a result, existing networks can only handle one specific image restoration application. To address such problems, we propose a novel adaptive residual network (ARN) for high-quality image restoration in this paper. Our ARN is a deep residual network, which is composed of convolutional layers, parametric rectified linear unit layers, and some adaptive shortcuts. We assign different scaling parameters to different inputs of the shortcuts, where the scaling is considered as part parameters of the ARN and trained adaptively according to different applications. Due to the special construction of ARN, it can solve many image restoration problems and have superior performance. We demonstrate its capabilities with three representative applications, including Gaussian image denoising, single image super resolution, and JPEG image deblocking. Experimental results prove that our model greatly outperforms numerous state-of-the-art restoration methods in terms of both peak signal-to-noise ratio and structure similarity index metrics, e.g., it achieves 0.2–0.3 dB gain in average compared with the second best method at a wide range of situations. [ABSTRACT FROM AUTHOR]

Crowd counting is a challenging task, mainly due to the severe occlusions among dense crowds. This paper aims to take a broader view to address crowd counting from the perspective of semantic modeling. In essence, crowd counting is a task of pedestrian semantic analysis involving three key factors: pedestrians, heads, and their context structure. The information of different body parts is an important cue to help us judge whether there exists a person at a certain position. Existing methods usually perform crowd counting from the perspective of directly modeling the visual properties of either the whole body or the heads only, without explicitly capturing the composite body-part semantic structure information that is crucial for crowd counting. In our approach, we first formulate the key factors of crowd counting as semantic scene models. Then, we convert the crowd counting problem into a multi-task learning problem, such that the semantic scene models are turned into different sub-tasks. Finally, the deep convolutional neural networks are used to learn the sub-tasks in a unified scheme. Our approach encodes the semantic nature of crowd counting and provides a novel solution in terms of pedestrian semantic analysis. In experiments, our approach outperforms the state-of-the-art methods on four benchmark crowd counting data sets. The semantic structure information is demonstrated to be an effective cue in scene of crowd counting. [ABSTRACT FROM PUBLISHER]

Deep neural networks have demonstrated remarkable reconstruction for single-image super-resolution (SISR). However, most existing CNN-based SISR methods directly learn the relation between low-resolution (LR) and high-resolution (HR) images, neglecting to explore the recurrence of internal patches, hence hindering the representational power of CNNs. In this paper, we propose a novel single image Super-Resolution network based on Graph ATtention network (SRGAT) to make full use of the internal patch-recurrence in a natural image. The proposed model employs a feature mapping block with a recurrent structure to refine low-level representations with high-level information. Especifically, the feature mapping block contains a parallel graph similarity branch and a content branch, where the graph similarity branch aims at exploiting the similarity and symmetry across different image patches in low-resolution feature space and provides additional priors for the content branch to enhance texture details. Specifically, we consider the internal patch-recurrence of an image by constructing a graph network on image feature patches. In this way, the information from neighboring patches can be interacted using graph attention network (GAT) to help it recover additional textures, which complements the textures learned from the content branch. Extensive quantitative and qualitative evaluations on five benchmark datasets demonstrate that the proposed algorithm performs favorably against the state-of-the-art super-resolution methods. [ABSTRACT FROM AUTHOR]

Although deep neural networks have achieved great success on numerous large-scale tasks, poor interpretability is still a notorious obstacle for practical applications. In this paper, we propose a novel and general attention mechanism, loss-based attention, upon which we modify deep neural networks to mine significant image patches for explaining which parts determine the image decision-making. This is inspired by the fact that some patches contain significant objects or their parts for image-level decision. Unlike previous attention mechanisms that adopt different layers and parameters to learn weights and image prediction, the proposed loss-based attention mechanism mines significant patches by utilizing the same parameters to learn patch weights and logits (class vectors), and image prediction simultaneously, so as to connect the attention mechanism with the loss function for boosting the patch precision and recall. Additionally, different from previous popular networks that utilize max-pooling or stride operations in convolutional layers without considering the spatial relationship of features, the modified deep architectures first remove them to preserve the spatial relationship of image patches and greatly reduce their dependencies, and then add two convolutional or capsule layers to extract their features. With the learned patch weights, the image-level decision of the modified deep architectures is the weighted sum on patches. Extensive experiments on large-scale benchmark databases demonstrate that the proposed architectures can obtain better or competitive performance to state-of-the-art baseline networks with better interpretability. The source codes are available on: https://github.com/xsshi2015/Loss-based-Attention-for-Interpreting-Image-level-Prediction-of-Convolutional-Neural-Networks. [ABSTRACT FROM AUTHOR]

Before the big data era, scene recognition was often approached with two-step inference using localized intermediate representations (objects, topics, and so on). One of such approaches is the semantic manifold (SM), in which patches and images are modeled as points in a semantic probability simplex. Patch models are learned resorting to weak supervision via image labels, which leads to the problem of scene categories co-occurring in this semantic space. Fortunately, each category has its own co-occurrence patterns that are consistent across the images in that category. Thus, discovering and modeling these patterns are critical to improve the recognition performance in this representation. Since the emergence of large data sets, such as ImageNet and Places, these approaches have been relegated in favor of the much more powerful convolutional neural networks (CNNs), which can automatically learn multi-layered representations from the data. In this paper, we address many limitations of the original SM approach and related works. We propose discriminative patch representations using neural networks and further propose a hybrid architecture in which the semantic manifold is built on top of multiscale CNNs. Both representations can be computed significantly faster than the Gaussian mixture models of the original SM. To combine multiple scales, spatial relations, and multiple features, we formulate rich context models using Markov random fields. To solve the optimization problem, we analyze global and local approaches, where a top–down hierarchical algorithm has the best performance. Experimental results show that exploiting different types of contextual relations jointly consistently improves the recognition accuracy. [ABSTRACT FROM AUTHOR]

Style transfer is a process of migrating a style from a given image to the content of another, synthesizing a new image, which is an artistic mixture of the two. Recent work on this problem adopting convolutional neural-networks (CNN) ignited a renewed interest in this field, due to the very impressive results obtained. There exists an alternative path toward handling the style transfer task, via the generalization of texture synthesis algorithms. This approach has been proposed over the years, but its results are typically less impressive compared with the CNN ones. In this paper, we propose a novel style transfer algorithm that extends the texture synthesis work of Kwatra et al. (2005), while aiming to get stylized images that are closer in quality to the CNN ones. We modify Kwatra’s algorithm in several key ways in order to achieve the desired transfer, with emphasis on a consistent way for keeping the content intact in selected regions, while producing hallucinated and rich style in others. The results obtained are visually pleasing and diverse, shown to be competitive with the recent CNN style transfer algorithms. The proposed algorithm is fast and flexible, being able to process any pair of content + style images. [ABSTRACT FROM PUBLISHER]

Visual saliency is a fundamental problem in both cognitive and computational sciences, including computer vision. In this paper, we discover that a high-quality visual saliency model can be learned from multiscale features extracted using deep convolutional neural networks (CNNs), which have had many successes in visual recognition tasks. For learning such saliency models, we introduce a neural network architecture, which has fully connected layers on top of CNNs responsible for feature extraction at three different scales. The penultimate layer of our neural network has been confirmed to be a discriminative high-level feature vector for saliency detection, which we call deep contrast feature. To generate a more robust feature, we integrate handcrafted low-level features with our deep contrast feature. To promote further research and evaluation of visual saliency models, we also construct a new large database of 4447 challenging images and their pixelwise saliency annotations. Experimental results demonstrate that our proposed method is capable of achieving the state-of-the-art performance on all public benchmarks, improving the F-measure by 6.12% and 10%, respectively, on the DUT-OMRON data set and our new data set (HKU-IS), and lowering the mean absolute error by 9% and 35.3%, respectively, on these two data sets. [ABSTRACT FROM PUBLISHER]

Multi-atlas parcellation (MAP) is carried out on a brain image by propagating and fusing labelled regions from brain atlases. Typical nonlinear registration-based label propagation is time-consuming and sensitive to inter-subject differences. Recently, deep learning parcellation (DLP) has been proposed to avoid nonlinear registration for better efficiency and robustness than MAP. However, most existing DLP methods neglect using brain atlases, which contain high-level information (e.g., manually labelled brain regions), to provide auxiliary features for improving the parcellation accuracy. In this paper, we propose a novel multi-atlas DLP method for brain parcellation. Our method is based on fully convolutional networks (FCN) and squeeze-and-excitation (SE) modules. It can automatically and adaptively select features from the most relevant brain atlases to guide parcellation. Moreover, our method is trained via a generative adversarial network (GAN), where a convolutional neural network (CNN) with multi-scale $l_{1}$ loss is used as the discriminator. Benefiting from brain atlases, our method outperforms MAP and state-of-the-art DLP methods on two public image datasets (LPBA40 and NIREP-NA0). [ABSTRACT FROM AUTHOR]

Object proposal is essential for current state-of-the-art object detection pipelines. However, the existing proposal methods generally fail in producing results with satisfying localization accuracy. The case is even worse for small objects, which, however, are quite common in practice. In this paper, we propose a novel scale-aware pixelwise object proposal network (SPOP-net) to tackle the challenges. The SPOP-net can generate proposals with high recall rate and average best overlap, even for small objects. In particular, in order to improve the localization accuracy, a fully convolutional network is employed which predicts locations of object proposals for each pixel. The produced ensemble of pixelwise object proposals enhances the chance of hitting the object significantly without incurring heavy extra computational cost. To solve the challenge of localizing objects at small scale, two localization networks, which are specialized for localizing objects with different scales are introduced, following the divide-and-conquer philosophy. Location outputs of these two networks are then adaptively combined to generate the final proposals by a large-/small-size weighting network. Extensive evaluations on PASCAL VOC 2007 and COCO 2014 show the SPOP network is superior over the state-of-the-art models. The high-quality proposals from SPOP-net also significantly improve the mean average precision of object detection with Fast-Regions with CNN features framework. Finally, the SPOP-net (trained on PASCAL VOC) shows great generalization performance when testing it on ILSVRC 2013 validation set. [ABSTRACT FROM PUBLISHER]

Recent deep learning models have demonstrated strong capabilities for classifying text and non-text components in natural images. They extract a high-level feature globally computed from a whole image component (patch), where the cluttered background information may dominate true text features in the deep representation. This leads to less discriminative power and poorer robustness. In this paper, we present a new system for scene text detection by proposing a novel text-attentional convolutional neural network (Text-CNN) that particularly focuses on extracting text-related regions and features from the image components. We develop a new learning mechanism to train the Text-CNN with multi-level and rich supervised information, including text region mask, character label, and binary text/non-text information. The rich supervision information enables the Text-CNN with a strong capability for discriminating ambiguous texts, and also increases its robustness against complicated background components. The training process is formulated as a multi-task learning problem, where low-level supervised information greatly facilitates the main task of text/non-text classification. In addition, a powerful low-level detector called contrast-enhancement maximally stable extremal regions (MSERs) is developed, which extends the widely used MSERs by enhancing intensity contrast between text patterns and background. This allows it to detect highly challenging text patterns, resulting in a higher recall. Our approach achieved promising results on the ICDAR 2013 data set, with an F-measure of 0.82, substantially improving the state-of-the-art results. [ABSTRACT FROM AUTHOR]

It is widely acknowledged that single image super-resolution (SISR) methods play a critical role in recovering the missing high-frequencies in an input low-resolution image. As SISR is severely ill-conditioned, image priors are necessary to regularize the solution spaces and generate the corresponding high-resolution image. In this paper, we propose an effective SISR framework based on the enhanced non-local similarity modeling and learning-based multi-directional feature prediction (ENLTV-MDFP). Since both the modeled and learned priors are exploited, the proposed ENLTV-MDFP method benefits from the complementary properties of the reconstruction-based and learning-based SISR approaches. Specifically, for the non-local similarity-based modeled prior [enhanced non-local total variation, (ENLTV)], it is characterized via the decaying kernel and stable group similarity reliability schemes. For the learned prior [multi-directional feature prediction prior, (MDFP)], it is learned via the deep convolutional neural network. The modeled prior performs well in enhancing edges and suppressing visual artifacts, while the learned prior is effective in hallucinating details from external images. Combining these two complementary priors in the MAP framework, a combined SR cost function is proposed. Finally, the combined SR problem is solved via the split Bregman iteration algorithm. Based on the extensive experiments, the proposed ENLTV-MDFP method outperforms many state-of-the-art algorithms visually and quantitatively. [ABSTRACT FROM AUTHOR]

Previous RGB-D fusion systems based on convolutional neural networks typically employ a two-stream architecture, in which RGB and depth inputs are learned independently. The multi-modal fusion stage is typically performed by concatenating the deep features from each stream in the inference process. The traditional two-stream architecture might experience insufficient multi-modal fusion due to two following limitations: 1) the cross-modal complementarity is rarely studied in the bottom–up path, wherein we believe the cross-modal complements can be combined to learn new discriminative features to enlarge the RGB-D representation community and 2) the cross-modal channels are typically combined by undifferentiated concatenation, which appears ambiguous to selecting cross-modal complementary features. In this paper, we address these two limitations by proposing a novel three-stream attention-aware multi-modal fusion network. In the proposed architecture, a cross-modal distillation stream, accompanying the RGB-specific and depth-specific streams, is introduced to extract new RGB-D features in each level in the bottom–up path. Furthermore, the channel-wise attention mechanism is innovatively introduced to the cross-modal cross-level fusion problem to adaptively select complementary feature maps from each modality in each level. Extensive experiments report the effectiveness of the proposed architecture and the significant improvement over the state-of-the-art RGB-D salient object detection methods. [ABSTRACT FROM AUTHOR]

Traditional saliency models usually adopt hand-crafted image features and human-designed mechanisms to calculate local or global contrast. In this paper, we propose a novel computational saliency model, i.e., deep spatial contextual long-term recurrent convolutional network (DSCLRCN), to predict where people look in natural scenes. DSCLRCN first automatically learns saliency related local features on each image location in parallel. Then, in contrast with most other deep network based saliency models which infer saliency in local contexts, DSCLRCN can mimic the cortical lateral inhibition mechanisms in human visual system to incorporate global contexts to assess the saliency of each image location by leveraging the deep spatial long short-term memory (DSLSTM) model. Moreover, we also integrate scene context modulation in DSLSTM for saliency inference, leading to a novel deep spatial contextual LSTM (DSCLSTM) model. The whole network can be trained end-to-end and works efficiently when testing. Experimental results on two benchmark datasets show that DSCLRCN can achieve state-of-the-art performance on saliency detection. Furthermore, the proposed DSCLSTM model can significantly boost the saliency detection performance by incorporating both global spatial interconnections and scene context modulation, which may uncover novel inspirations for studies on them in computational saliency models. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (CNNs) have shown their great success on image classification. CNNs mainly consist of convolutional and pooling layers, both of which are performed on local image areas without considering the dependence among different image regions. However, such dependence is very important for generating explicit image representation. In contrast, recurrent neural networks (RNNs) are well known for their ability of encoding contextual information in sequential data, and they only require a limited number of network parameters. Thus, we proposed the hierarchical RNNs (HRNNs) to encode the contextual dependence in image representation. In HRNNs, each RNN layer focuses on modeling spatial dependence among image regions from the same scale but different locations. While the cross RNN scale connections target on modeling scale dependencies among regions from the same location but different scales. Specifically, we propose two RNN models: 1) hierarchical simple recurrent network (HSRN), which is fast and has low computational cost and 2) hierarchical long-short term memory recurrent network, which performs better than HSRN with the price of higher computational cost. In this paper, we integrate CNNs with HRNNs, and develop end-to-end convolutional hierarchical RNNs (C-HRNNs) for image classification. C-HRNNs not only utilize the discriminative representation power of CNNs, but also utilize the contextual dependence learning ability of our HRNNs. On four of the most challenging object/scene image classification benchmarks, our C-HRNNs achieve the state-of-the-art results on Places 205, SUN 397, and MIT indoor, and the competitive results on ILSVRC 2012. [ABSTRACT FROM PUBLISHER]

The performance of object detection has recently been significantly improved due to the powerful features learnt through convolutional neural networks (CNNs). Despite the remarkable success, there are still several major challenges in object detection, including object rotation, within-class diversity, and between-class similarity, which generally degenerate object detection performance. To address these issues, we build up the existing state-of-the-art object detection systems and propose a simple but effective method to train rotation-invariant and Fisher discriminative CNN models to further boost object detection performance. This is achieved by optimizing a new objective function that explicitly imposes a rotation-invariant regularizer and a Fisher discrimination regularizer on the CNN features. Specifically, the first regularizer enforces the CNN feature representations of the training samples before and after rotation to be mapped closely to each other in order to achieve rotation-invariance. The second regularizer constrains the CNN features to have small within-class scatter but large between-class separation. We implement our proposed method under four popular object detection frameworks, including region-CNN (R-CNN), Fast R- CNN, Faster R- CNN, and R- FCN. In the experiments, we comprehensively evaluate the proposed method on the PASCAL VOC 2007 and 2012 data sets and a publicly available aerial image data set. Our proposed methods outperform the existing baseline methods and achieve the state-of-the-art results. [ABSTRACT FROM AUTHOR]

Image matting is widely studied for accurate foreground extraction. Most algorithms, including deep-learning based solutions, require a carefully edited trimap. Recent works attempt to combine the segmentation stage and matting stage in one CNN model, but errors occurring at the segmentation stage lead to unsatisfactory matte. We propose a user-guided approach for practical human matting. More precisely, we provide a good automatic initial matting and a natural way of interaction that reduces the workload of drawing trimaps and allows users to guide the matting in ambiguous situation. We also combine the segmentation and matting stage in an end-to-end CNN architecture and introduce a residual-learning module to support convenient stroke-based interaction. The proposed model learns to propagate the input trimap and modify the deep image features, which can efficiently correct the segmentation errors. Our model supports arbitrary forms of trimaps from carefully edited to totally unknown maps. Our model also allows users to choose from different foreground estimations according to their preference. We collected a large human matting dataset consisting of 12K real-world human images with complex background and human-object relations. The proposed model is trained on the new dataset with a novel trimap generation strategy that enables the model to tackle different test situations and highly improves the interaction efficiency. Our method outperforms other state-of-the-art automatic methods and achieve competitive accuracy when high-quality trimaps are provided. Experiments indicate that our interactive matting strategy is superior to separately estimating the trimap and alpha matte using two models. [ABSTRACT FROM AUTHOR]

The goal of exemplar-based texture synthesis is to generate texture images that are visually similar to a given exemplar. Recently, promising results have been reported by methods relying on convolutional neural networks (ConvNets) pretrained on large-scale image datasets. However, these methods have difficulties in synthesizing image textures with non-local structures and extending to dynamic or sound textures. In this article, we present a conditional generative ConvNet (cgCNN) model which combines deep statistics and the probabilistic framework of generative ConvNet (gCNN) model. Given a texture exemplar, cgCNN defines a conditional distribution using deep statistics of a ConvNet, and synthesizes new textures by sampling from the conditional distribution. In contrast to previous deep texture models, the proposed cgCNN does not rely on pre-trained ConvNets but learns the weights of ConvNets for each input exemplar instead. As a result, cgCNN can synthesize high quality dynamic, sound and image textures in a unified manner. We also explore the theoretical connections between our model and other texture models. Further investigations show that the cgCNN model can be easily generalized to texture expansion and inpainting. Extensive experiments demonstrate that our model can achieve better or at least comparable results than the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Street Scene Change Detection (SSCD) aims to locate the changed regions between a given street-view image pair captured at different times, which is an important yet challenging task in the computer vision community. The intuitive way to solve the SSCD task is to fuse the extracted image feature pairs, and then directly measure the dissimilarity parts for producing a change map. Therefore, the key for the SSCD task is to design an effective feature fusion method that can improve the accuracy of the corresponding change maps. To this end, we present a novel Hierarchical Paired Channel Fusion Network (HPCFNet), which utilizes the adaptive fusion of paired feature channels. Specifically, the features of a given image pair are jointly extracted by a Siamese Convolutional Neural Network (SCNN) and hierarchically combined by exploring the fusion of channel pairs at multiple feature levels. In addition, based on the observation that the distribution of scene changes is diverse, we further propose a Multi-Part Feature Learning (MPFL) strategy to detect diverse changes. Based on the MPFL strategy, our framework achieves a novel approach to adapt to the scale and location diversities of the scene change regions. Extensive experiments on three public datasets (i.e., PCD, VL-CMU-CD and CDnet2014) demonstrate that the proposed framework achieves superior performance which outperforms other state-of-the-art methods with a considerable margin. [ABSTRACT FROM AUTHOR]

We propose an effective method to boost the accuracy of multi-person pose estimation in images. Initially, the three-layer fractal network was constructed to regress multi-person joints location heatmap that can help to enhance an image region with receptive field and capture more joints local-contextual feature information, thereby producing keypoints heatmap intermediate prediction to optimize human body joints regression results. Subsequently, the hierarchical bi-directional inference algorithm was proposed to calculate the degree of relatedness (call it Kinship) for adjacent joints, and it combines the Kinship between adjacent joints with the spatial constraints, which we refer to as joints kinship pattern matching mechanism, to determine the best matched joints pair. We iterate the above-mentioned joints matching process layer by layer until all joints are assigned to a corresponding individual. Comprehensive experiments demonstrate that the proposed approach outperforms the state-of-the-art schemes and achieves about 1% and 0.6% increase in mAP on MPII multi-person subset and MSCOCO 2016 keypoints challenge. [ABSTRACT FROM AUTHOR]

The wide 3D applications have led to increasing amount of 3D object data, and thus effective 3D object classification technique has become an urgent requirement. One important and challenging task for 3D object classification is how to formulate the 3D data correlation and exploit it. Most of the previous works focus on learning optimal pairwise distance metric for object comparison, which may lose the global correlation among 3D objects. Recently, a transductive hypergraph learning has been investigated for classification, which can jointly explore the correlation among multiple objects, including both the labeled and unlabeled data. Although these methods have shown better performance, they are still limited due to 1) a considerable amount of testing data may not be available in practice and 2) the high computational cost to test new coming data. To handle this problem, considering the multi-modal representations of 3D objects in practice, we propose an inductive multi-hypergraph learning algorithm, which targets on learning an optimal projection for the multi-modal training data. In this method, all the training data are formulated in multi-hypergraph based on the features, and the inductive learning is conducted to learn the projection matrices and the optimal multi-hypergraph combination weights simultaneously. Different from the transductive learning on hypergraph, the high cost training process is off-line, and the testing process is very efficient for the inductive learning on hypergraph. We have conducted experiments on two 3D benchmarks, i.e., the NTU and the ModelNet40 data sets, and compared the proposed algorithm with the state-of-the-art methods and traditional transductive multi-hypergraph learning methods. Experimental results have demonstrated that the proposed method can achieve effective and efficient classification performance. We also note that the proposed method is a general framework and has the potential to be applied in other applications in practice. [ABSTRACT FROM AUTHOR]