Search

In this paper, we conduct an extensive study on the use of pre-trained convolutional neural networks (CNNs) for omnidirectional image quality assessment (IQA). To cope with the lack of available IQA databases, transfer learning from seven pre-trained CNN models is investigated over retraining on standard 2D databases. In addition, we explore the influence of various image representations and training strategies on the model’s performance. A comparison of the use of projected versus radial content, and multichannel CNN versus patch-wise training is also covered. The experimental results on two publicly available databases are used to draw conclusions about which strategy best fits the visual quality prediction and at which computational cost. The analysis shows that retraining CNN models on 2D IQA databases improves the prediction accuracy. The latter and the required computational time are found to be significantly affected by the training strategy. Cross-database evaluations demonstrate that the nature and variety of the content impact the generalization ability of the models. Finally, we show that conclusions coming from other image processing communities may not hold for IQA. The provided discussion shall provide insights and recommendations when using pre-trained CNNs for omnidirectional IQA. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel deep learning-based visual comfort assessment (VCA) for stereoscopic images. To assess the overall degree of visual discomfort in stereoscopic viewing, we devise a binocular fusion deep network (BFN) learning binocular characteristics between stereoscopic images. The proposed BFN learns the latent binocular feature representations for the visual comfort score prediction. In the BFN, the binocular feature is encoded by fusing the spatial features extracted from left and right views. Finally, the visual comfort score is predicted by projecting the binocular feature onto the subjective score space. In addition, we devise a disparity regularization network (DRN) for improving the prediction results. The proposed DRN takes the binocular feature from the BFN and estimates disparity maps from the feature in order to embed disparity relations between left and right views into the deep network. The proposed deep network with BFN and DRN is end-to-end trained in a unified framework in which the DRN acts as disparity regularization. We evaluated the prediction performance of the proposed deep network for VCA by the comparison of existing objective VCA metrics. Further, we demonstrated that the proposed BFN showed various factors causing visual discomfort by using network visualization. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have recently shown excellent performance on the task of fine-grained vehicle classification, where the motivation is to identify the fine-grained categories of the given vehicles. Generally speaking, the main motivation of the conventional back-propagation algorithm is to optimize the loss function. The algorithm itself does not guarantee if the extracted features are discriminative for the task of classification. Intuitively, if we can learn more discriminative features with a relatively small number of feature maps, the generalization ability of the CNNs will be significantly improved. Therefore, we propose a channel max pooling (CMP) scheme, where a new layer is inserted between the fully connected layers and the convolutional layers. The proposed CMP scheme divides the feature maps into to several sub-groups. Then, it compresses the feature maps within each sub-group into a new one. The compression is carried out by selecting the maximum value among the same locations from different feature maps. Moreover, the proposed CMP layer has the advantage that it can reduce the number of parameters via reducing the number of channels in the CNNs. Experimental results on two fine-grained vehicle datasets demonstrate that the CMP modified CNNs can improve the classification accuracies on the task of fine-grained vehicle classification while a massive amount of parameters are reduced. Moreover, it has competitive performance when comparing with the-state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The main challenges of visual object tracking arise from the arbitrary appearance of the objects that need to be tracked. Most existing algorithms try to solve this problem by training a new model to regenerate or classify each tracked object. As a result, the model needs to be initialized and retrained for each new object. In this paper, we propose to track different objects in an object-independent approach with a novel two-flow convolutional neural network (YCNN). The YCNN takes two inputs (one is an object image patch, the other is a larger searching image patch), then outputs a response map which predicts how likely and where the object would appear in the search patch. Unlike the object-specific approaches, the YCNN is actually trained to measure the similarity between the two image patches. Thus, this model will not be limited to any specific object. Furthermore, the network is end-to-end trained to extract both shallow and deep dedicated convolutional features for visual tracking. And once properly trained, the YCNN can be used to track all kinds of objects without further training and updating. As a result, our algorithm is able to run at a very high speed of 45 frames-per-second. The effectiveness of the proposed algorithm can also be proved by the experiments on two popular data sets: OTB-100 and VOT-2014. [ABSTRACT FROM AUTHOR]

Individuating and locating repetitive patterns in still images is a fundamental task in image processing, typically achieved by means of correlation strategies. In this paper, we provide a solid solution to this task using a differential geometry approach, operating on Lie algebra, and exploiting a mixture of templates. The proposed method asks the user to locate a few instances of the target patterns (seeds) that become visual templates used to explore the image. We propose an iterative algorithm to locate patches similar to the seeds working in three steps: first, clustering the detected patches to generate templates of different classes, then looking for the affine transformations, living on a Lie algebra that best links the templates and the detected patches, and finally detecting new patches with a convolutional strategy. The process ends when no new patches are found. We will show how our method is able to process heterogeneous unstructured images with multiple visual motifs and extremely crowded scenarios with high precision and recall, outperforming all the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper aims for event recognition when video examples are scarce or even completely absent. The key in such a challenging setting is a semantic video representation. Rather than building the representation from individual attribute detectors and their annotations, we propose to learn the entire representation from freely available web videos and their descriptions using an embedding between video features and term vectors. In our proposed embedding, which we call Video2vec, the correlations between the words are utilized to learn a more effective representation by optimizing a joint objective balancing descriptiveness and predictability. We show how learning the Video2vec embedding using a multimodal predictability loss, including appearance, motion and audio features, results in a better predictable representation. We also propose an event specific variant of Video2vec to learn a more accurate representation for the words, which are indicative of the event, by introducing a term sensitive descriptiveness loss. Our experiments on three challenging collections of web videos from the NIST TRECVID Multimedia Event Detection and Columbia Consumer Videos datasets demonstrate: i) the advantages of Video2vec over representations using attributes or alternative embeddings, ii) the benefit of fusing video modalities by an embedding over common strategies, iii) the complementarity of term sensitive descriptiveness and multimodal predictability for event recognition. By its ability to improve predictability of present day audio-visual video features, while at the same time maximizing their semantic descriptiveness, Video2vec leads to state-of-the-art accuracy for both few- and zero-example recognition of events in video. [ABSTRACT FROM PUBLISHER]

The use of local features for image representation has become popular in recent years. Local features are often used in the bag-of-visual-words scheme. Although proven effective, this method still has two drawbacks. First, local regions from which local features are extracted are not discriminative enough for visual tasks. Hence, the combination of local features is necessary. Second, the semantic gap between visual features and human perception also hinders the performance. To address these two problems, in this paper, we propose a novel contextual exemplar classifier-based method for image representation and apply it for classification tasks. Each exemplar classifier is trained to separate one training image from the other images of different classes. We partition each image into a number of regions and use the responses of these exemplar classifiers as the image region’s representation. The contextual relationship is then modeled using mixture Dirichlet distributions. A bilayer model is used to predict image classes with L2 constraints. Experimental results on the Natural Scene, Caltech-101/256, Flower-17/102, and SUN-397 data sets show that the proposed method is able to outperform the state-of-the-art local feature-based methods for image classification. [ABSTRACT FROM PUBLISHER]

We present a framework to automatically detect and remove shadows in real world scenes from a single image. Previous works on shadow detection put a lot of effort in designing shadow variant and invariant hand-crafted features. In contrast, our framework automatically learns the most relevant features in a supervised manner using multiple convolutional deep neural networks (ConvNets). The features are learned at the super-pixel level and along the dominant boundaries in the image. The predicted posteriors based on the learned features are fed to a conditional random field model to generate smooth shadow masks. Using the detected shadow masks, we propose a Bayesian formulation to accurately extract shadow matte and subsequently remove shadows. The Bayesian formulation is based on a novel model which accurately models the shadow generation process in the umbra and penumbra regions. The model parameters are efficiently estimated using an iterative optimization procedure. Our proposed framework consistently performed better than the state-of-the-art on all major shadow databases collected under a variety of conditions. [ABSTRACT FROM PUBLISHER]

Localizing objects of interest in images when provided with only image-level labels is a challenging visual recognition task. Previous efforts have required carefully designed features and have difficulty in handling images with cluttered backgrounds. Up-scaling to large datasets also poses a challenge to applying these methods to real applications. In this paper, we propose an efficient and effective learning framework called MILinear, which is able to learn an object localization model from large-scale data without using bounding box annotations. We integrate rich general prior knowledge into a learning model using a large pre-trained convolutional network. Moreover, to reduce ambiguity in positive images, we present a bag-splitting algorithm that iteratively generates new negative bags from positive ones. We evaluate the proposed approach on the challenging Pascal VOC 2007 dataset, and our method outperforms other state-of-the-art methods by a large margin; some results are even comparable to fully supervised models trained with bounding box annotations. To further demonstrate scalability, we also present detection results on the ILSVRC 2013 detection dataset, and our method outperforms supervised deformable part-based model without using box annotations. [ABSTRACT FROM PUBLISHER]

Gait recognition is a useful biometric trait for person authentication because it is usable even with low image resolution. One challenge is robustness to a view change (cross-view matching); view transformation models (VTMs) have been proposed to solve this. The VTMs work well if the target views are the same as their discrete training views. However, the gait traits are observed from an arbitrary view in a real situation. Thus, the target views may not coincide with discrete training views, resulting in recognition accuracy degradation. We propose an arbitrary VTM (AVTM) that accurately matches a pair of gait traits from an arbitrary view. To realize an AVTM, we first construct 3D gait volume sequences of training subjects, disjoint from the test subjects in the target scene. We then generate 2D gait silhouette sequences of the training subjects by projecting the 3D gait volume sequences onto the same views as the target views, and train the AVTM with gait features extracted from the 2D sequences. In addition, we extend our AVTM by incorporating a part-dependent view selection scheme (AVTM_PdVS), which divides the gait feature into several parts, and sets part-dependent destination views for transformation. Because appropriate destination views may differ for different body parts, the part-dependent destination view selection can suppress transformation errors, leading to increased recognition accuracy. Experiments using data sets collected in different settings show that the AVTM improves the accuracy of cross-view matching and that the AVTM_PdVS further improves the accuracy in many cases, in particular, verification scenarios. [ABSTRACT FROM AUTHOR]

This paper presents a new automatic image annotation algorithm. First, we introduce a new similarity measure between images: compactness. This uses low level visual descriptors for determining the similarity between two images. Compactness shows how close test image features lie to training image feature cluster centers. The measure provides the core for a k-nearest neighbor type image annotation method. Afterward, a formalism for defining different transfer techniques is devised and several label transfer techniques are provided. The method as whole is evaluated on four image annotation benchmarks. The results on these sets validate the accuracy of the approach, which outperforms many state-of-the-art annotation methods. The method presented here requires a simple training process, efficiently combines different feature types and performs better than complex learning algorithms, even in this incipient form. The main contributions of this paper are the usage of compactness as a similarity measure that enables efficient low level feature comparison and an annotation algorithm based on label transfer. [ABSTRACT FROM AUTHOR]

Medical image analysis remains a challenging application area for artificial intelligence. When applying machine learning, obtaining ground-truth labels for supervised learning is more difficult than in many more common applications of machine learning. This is especially so for datasets with abnormalities, as tissue types and the shapes of the organs in these datasets differ widely. However, organ detection in such an abnormal dataset may have many promising potential real-world applications, such as automatic diagnosis, automated radiotherapy planning, and medical image retrieval, where new multimodal medical images provide more information about the imaged tissues for diagnosis. Here, we test the application of deep learning methods to organ identification in magnetic resonance medical images, with visual and temporal hierarchical features learned to categorize object classes from an unlabeled multimodal DCE-MRI dataset so that only a weakly supervised training is required for a classifier. A probabilistic patch-based method was employed for multiple organ detection, with the features learned from the deep learning model. This shows the potential of the deep learning model for application to medical images, despite the difficulty of obtaining libraries of correctly labeled training datasets and despite the intrinsic abnormalities present in patient datasets. [ABSTRACT FROM PUBLISHER]

Measuring image similarity is a central topic in computer vision. In this paper, we propose to measure image similarity by learning from the online Flickr image groups. We do so by: Choosing 103 Flickr groups, building a one-versus-all multiclass classifier to classify test images into a group, taking the set of responses of the classifiers as features, calculating the distance between feature vectors to measure image similarity. Experimental results on the Corel dataset and the PASCAL VOC 2007 dataset show that our approach performs better on image matching, retrieval, and classification than using conventional visual features. To build our similarity measure, we need one-versus-all classifiers that are accurate and can be trained quickly on very large quantities of data. We adopt an SVM classifier with a histogram intersection kernel. We describe a novel fast training algorithm for this classifier: the Stochastic Intersection Kernel MAchine (SIKMA) training algorithm. This method can produce a kernel classifier that is more accurate than a linear classifier on tens of thousands of examples in minutes. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive recognition model (ARM) is proposed for image annotation. The ARM consists of an adaptive classification network (CFN) and a nonlinear correlation network (CLN). The adaptive CFN aims to annotate an image with keywords, and the CLN is used to unveil the correlative information of keywords for annotation refinement. Image annotation is carried out by an ARM in two stages. In the first stage, the features extracted from regions of the input image are fed to a CFN to produce classification labels. In the second stage, the CLN uses keyword correlations learned from the training images to refine the classification result. The ARM works in a forward-propagating manner, resulting in high efficiency in image annotation. Furthermore, the computational time of an ARM is insensitive to the number of regions of the input image and the vocabulary size. In this paper, the effect of keyword correlation in image annotation is, comprehensively, investigated on a real image dataset and a synthetic image dataset. The exploitation of a controllable synthetic dataset helps to systematically study the function of keyword correlation and effectively analyze the performance of the ARM. Experimental results demonstrate the efficiency and effectiveness of the ARM. [ABSTRACT FROM PUBLISHER]

Automatic image annotation (AIA) is very important to image retrieval and image understanding. Two key issues in AIA are explored in detail in this paper, i.e., structured visual feature selection and the implementation of hierarchical correlated structures among multiple tags to boost the performance of image annotation. This paper simultaneously introduces an input and output structural grouping sparsity into a regularized regression model for image annotation. For input high-dimensional heterogeneous features such as color, texture, and shape, different kinds (groups) of features have different intrinsic discriminative power for the recognition of certain concepts. The proposed structured feature selection by structural grouping sparsity can be used not only to select group-of-features but also to conduct within-group selection. Hierarchical correlations among output labels are well represented by a tree structure, and therefore, the proposed tree-structured grouping sparsity can be used to boost the performance of multitag image annotation. In order to efficiently solve the proposed regression model, we relax the solving process as a framework of the bilayer regression model for multilabel boosting by the selection of heterogeneous features with structural grouping sparsity (Bi-MtBGS). The first-layer regression is to select the discriminative features for each label. The aim of the second-layer regression is to refine the feature selection model learned from the first layer, which can be taken as a multilabel boosting process. Extensive experiments on public benchmark image data sets and real-world image data sets demonstrate that the proposed approach has better performance of multitag image annotation and leads to a quite interpretable model for image understanding. [ABSTRACT FROM AUTHOR]

Visual attention, which is an important characteristic of human visual system, is a useful clue for image processing and compression applications in the real world. This paper proposes a computational scheme that adopts both low-level and high-level features to predict visual attention from video signal by machine learning. The adoption of low-level features (color, orientation, and motion) is based on the study of visual cells, and the adoption of the human face as a high-level feature is based on the study of media communications. We show that such a scheme is more robust than those using purely single low- or high-level features. Unlike conventional techniques, our scheme is able to learn the relationship between features and visual attention to avoid perceptual mismatch between the estimated salience and the actual human fixation. We also show that selecting the representative training samples according to the fixation distribution improves the efficacy of regressive training. Experimental results are shown to demonstrate the advantages of the proposed scheme. [ABSTRACT FROM PUBLISHER]

In this paper, a structured max-margin learning algorithm is developed to achieve more effective training of a large number of inter-related classifiers for multilabel image annotation application. To leverage multilabel images for classifier training, each multilabel image is partitioned into a set of image instances (image regions or image patches) and an automatic instance label identification algorithm is developed to assign multiple labels (which are given at the image level) to the most relevant image instances. A K-way min-max cut algorithm is developed for automatic instance clustering and kernel weight determination, where multiple base kernels are seamlessly combined to address the issue of huge intra-concept visual diversity more effectively. Second, a visual concept network is constructed for characterizing the inter-concept visual similarity contexts more precisely in the high-dimensional multimodal feature space. The visual concept network is used to determine the inter-related learning tasks directly in the feature space rather than in the label space because feature space is the common space for classifier training and image classification. Third, a parallel computing platform is developed to achieve more effective learning of a large number of inter-related classifiers over the visual concept network. A structured max-margin learning algorithm is developed by incorporating the visual concept network, max-margin Markov networks and multitask learning to address the issue of huge inter-concept visual similarity more effectively. By leveraging the inter-concept visual similarity contexts for inter-related classifier training, our structured max-margin learning algorithm can significantly enhance the discrimination power of the inter-related classifiers. Our experiments have also obtained very positive results for a large number of object classes and image concepts. [ABSTRACT FROM AUTHOR]