Search

Existing video summarization approaches mainly concentrate on the sequential or structural characteristic of video data. However, they do not pay enough attention to the video summarization task itself. In this article, we propose a meta learning method for performing task-driven video summarization, denoted by MetaL-TDVS, to explicitly explore the video summarization mechanism among summarizing processes on different videos. Particularly, MetaL-TDVS aims to excavate the latent mechanism for summarizing video by reformulating video summarization as a meta learning problem and promote the generalization ability of the trained model. MetaL-TDVS regards summarizing each video as a single task to make better use of the experience and knowledge learned from processes of summarizing other videos to summarize new ones. Furthermore, MetaL-TDVS updates models via a twofold backpropagation, which forces the model optimized on one video to obtain high accuracy on another video in every training step. Extensive experiments on benchmark datasets demonstrate the superiority and better generalization ability of MetaL-TDVS against several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Deep learning-based hyperspectral image superresolution methods have achieved great success recently. However, most methods utilize 2D or 3D convolution to explore features, and rarely combine the two types of convolution to design networks. Moreover, when the model only contains 3D convolution, almost all the methods take all the bands of hyperspectral image as input to analyze, which requires more memory footprint. To address these issues, we explore a new structure for hyperspectral image superresolution using spectrum and feature context. Inspired by the high similarity among adjacent bands, we design a dual-channel network through 2D and 3D convolution to jointly exploit the information from both single band and adjacent bands, which is different from previous works. Under the connection of depth split, it can effectively share spatial information so as to improve the learning ability of 2D spatial domain. Besides, our method introduces the features extracted from previous band, which contributes to the complementarity of information and simplifies the network structure. Through feature context fusion, it significantly enhances the performance of the algorithm. Extensive evaluations and comparisons on three public datasets demonstrate that our approach produces the state-of-the-art results over the existing approaches. [ABSTRACT FROM AUTHOR]

Motion is an important clue for industrial inspection, video surveillance, and service machines to localize and recognize products and objects. Because blur co-occurs with motion, it is desirable for developing efficient and robust motion blur detection algorithm. However, existing algorithms are inefficient for detecting spatially varying motion blur. To deal with the problem, this paper presents a theorem, according to which, motion blur can be efficiently detected and segmented. According to the theorem, the proposed algorithm requires a simple filtering operation and variance computation. Classification as either blurred or unblurred pixel can be done by substituting the variance into the proposed simple formula and checking the sign of the resulting value. Moreover, a geometric interpretation and two extensions of the algorithm are given. Importantly, based on the geometric interpretation of the indicator function, we develop a one-class classifier, which is more effective than the indicator function and has comparable computational cost of the indicator function. Experimental results on detecting motion-blurred cars, motorcycles, bicycles, bags, and persons demonstrate that the proposed algorithm is very efficient without loss of effectiveness. [ABSTRACT FROM PUBLISHER]

Dimensionality reduction has attracted much research interest in the past few decades. Existing dimensionality reduction methods like linear discriminant analysis and principal component analysis have achieved promising performance, but the single and linear projection properties limit further improvements of performance. A novel convolutional two-dimensional nonlinear discriminant analysis method is proposed for dimensionality reduction in this article. In order to handle nonlinear data properly, we present a newly designed structure with convolutional neural networks (CNNs) to realize an equivalent objective function with classical two-dimensional linear discriminant analysis (2DLDA) and thus embed the original 2DLDA into an end-to-end network. In this way, the proposed dimensionality reduction network can utilize the nonlinearity of the CNN and benefit from the learning ability. The results of experiment on different image-related applications demonstrate that our method outperforms other comparable approaches, and its effectiveness is proved. [ABSTRACT FROM AUTHOR]

To fill the semantic gap between the predictive power of computational saliency models and human behavior, this paper proposes to predict where people look at using spatial-aware object-level cues. While object-level saliency has been recently suggested by psychophysics experiments and shown effective with a few computational models, the spatial relationship between the objects has not yet been explored in this context. We in this work for the first time explicitly model such spatial relationship, as well as leveraging semantic information of an image to enhance object-level saliency modeling. The core computational module is a graphlet-based (i.e., graphlets are moderate-sized connected subgraphs) deep architecture, which hierarchically learns a saliency map from raw image pixels to object-level graphlets (oGLs) and further to spatial-level graphlets (sGLs). Eye tracking data are also used to leverage human experience in saliency prediction. Experimental results demonstrate that the proposed oGLs and sGLs well capture object-level and spatial-level cues relating to saliency, and the resulting saliency model performs competitively compared with the state-of-the-art. [ABSTRACT FROM PUBLISHER]

In this paper, a new fine-grained image categorization system that improves spatial pyramid matching is developed. In this method, multiple cells are combined into cellets in the proposed categorization model, which are highly responsive to an object's fine categories. The object components are represented by cellets that can connect spatially adjacent cells within the same pyramid level. Here, image categorization can be formulated as the matching between the cellets of corresponding images. Toward an effective matching process, an active learning algorithm that can effectively select a few representative cells for constructing the cellets is adopted. A linear-discriminant-analysis-like scheme is employed to select discriminative cellets. Then, fine-grained image categorization can be conducted with a trained linear support vector machine. Experimental results on three real-world data sets demonstrate that our proposed system outperforms the state of the art. [ABSTRACT FROM PUBLISHER]