Understanding dynamic textures (DTs) is a challenge in various computer vision applications due to the negative impacts of noise, changes of environment, illumination, and scales on capturing turbulent characteristics. In this work, we propose an efficient shallow framework for DT representation by addressing the following novel concepts. First, it is the first time in DT analysis that 2D/3D Gaussian-gradient filterings are taken into account as a pre-processing step to point out robust components against those influences in effect. Second, high-order partial derivatives of the Gaussian kernels and their informative magnitudes are exploited to forcefully capture multi-order Gaussian-gradient features. Third, these gradient kernels are investigated in multi-scale analysis of different orders and standard deviations in order to enrich more useful scale-gradient information. Finally, the obtained complementary components are shallowly encoded using a simple local operator to construct robust descriptors of High-order 2D/3D Gaussian-gradient-based Features ($\mathrm{HoGF}^{2D/3D}$) against the well-known issues of DT description. Experiments for DT classification on various benchmarks have validated the interest of our approach since its performance is comparable to state-of-the-art results, including that of deep-learning methods, while it only has a small dimension.

Representation of dynamic textures (DTs), well‐known as a sequence of moving textures, is a challenging problem in video analysis due to the disorientation of motion features. Analysing DTs to make them ‘understandable’ plays an important role in different applications of computer vision. In this study, an efficient approach for DT description is proposed by addressing the following novel concepts. First, the beneficial properties of dense trajectories are exploited for the first time to efficiently describe DTs instead of the whole video. Second, two substantial extensions of local vector pattern operator are introduced to form a completed model which is based on complemented components to enhance its performance in encoding directional features of motion points in a trajectory. Finally, the authors present a new framework, called directional dense trajectory patterns, which takes advantage of directional beams of dense trajectories along with spatio‐temporal features of their motion points in order to construct dense‐trajectory‐based descriptors with more robustness. Evaluations of DT recognition on different benchmark datasets (i.e. UCLA, DynTex, and DynTex++) have verified the interest of the authors’ proposal.

An efficient framework for dynamic texture (DT) representation is proposed by exploiting local features based on Local Binary Patterns (LBP) from filtered images. First, Gaussian smoothing filter is used to deal with near uniform regions and noise which are typical restrictions of LBP operator. Second, the receptive field of Difference of Gaussians (DoG), which is exploited in DT description for the first time, allows to make the descriptor more robust against the changes of environment, illumination, and scale which are main challenges in DT representation. Experimental results of DT recognition on different benchmark datasets (i.e., UCLA, DynTex, and DynTex++), which give outstanding performance compared to the state of the art, verify the interest of our proposal.

Situation awareness in maritime environment entails early detection and classification of maritime targets of varying sizes, depths, shapes, textures, and contrasts. Thus, this paper describes a novel deep learning based maritime situation awareness approach using high-definition video. Maritime object detection is achieved in three main steps. At first, a key region based tracking algorithm allows to, dynamically and parsimoniously, extract high-quality region proposals mainly focalized around rigid (i.e.; potential object) video locations. The latter are, further, fed into a Fast-RCNN for carrying out objectness detection and box regression. Finally, a mere box post-regression operation enables to extract maritime objects. Furthermore, the found object detections are fed into a second classification RCNN, specifically, trained to recognize up to 40 vessel classes. Our experiments have shown that the proposed approach achieves state of the art speed and accuracy.

Reducing the amount of keypoints used to index an image is particularly interesting to control processing time and memory usage in real-time document image matching applications, like augmented documents or smartphone applications. This paper benchmarks two keypoint selection methods on a task consisting of reducing keypoint sets extracted from document images, while preserving detection and segmentation accuracy. We first study the different forms of keypoint filtering, and we introduce the use of the CORE selection method on keypoints extracted from document images. Then, we extend a previously published benchmark by including evaluations of the new method, by adding the SURF-BRISK detection/description scheme, and by reporting processing speeds. Evaluations are conducted on the publicly available dataset of ICDAR2015 SmartDOC challenge 1. Finally, we prove that reducing the original keypoint set is always feasible and can be beneficial not only to processing speed but also to accuracy.

This paper describes a novel approach for rigid object segmentation from a dynamic background using a pre-recorded video with a moving camera, and we apply it to the problem of vessel segmentation in a maritime video scene. The difficulty of modeling background appearance or/and dynamic, modeling object appearance, and compensating camera motion renders this task very challenging. Therefore, the proposed method only uses a geometric constraint regarding target motion rigidity in order to achieve object segmentation in a full video segment. Such an idea is not new in object segmentation literature, but the novelty in this paper resides in that the target rigidity assumption is implemented at the pixel level, but not at the object scale. This is firstly achieved by deriving a theoretical optic flow model in the neighborhood of each pixel under the 3D rigid motion assumption of object, which is later compared against the observed 2D optic flow model in the neighborhood of a pixel in order to derive a pixelwise rigidity criterion. The latter is further reenforced along individual (pixel) trajectories in a video segment. Finally, a mere thresholding operation allows to quickly extract target from background in a full video segment. Our experiments using real maritime video sequences captured with an airborne camera have shown that the method detects maritime targets accurately.

We present, in this paper, the FAIR algorithm: a fast algorithm for document image restoration. This algorithm has been submitted to different contests where it showed good performance in comparison to the state of the art. In addition, this method is scale invariant and fast enough to be used in real-time applications. The method is based on a double-threshold edge detection approach that makes it possible to detect small details while remaining robust against noise. The performance of the proposition is evaluated on several types of degraded document images where considerable background noise or variation in contrast and illumination exist.

In this paper, we present a novel approach for super-resolved binarization of document images acquired by low quality devices. The algorithm tries to compute the super resolution of the likelihood of text instead of the gray value of pixels. This method is the extension of a binarization algorithm (FAIR: a Fast Algorithm for document Image Restoration) which has been submitted into different contests where it showed good performances. The method can be considered as parameter free and is based on a rough localization of text in order to save computation time. Experimental results on several image sequences presenting a background noise and variation in contrast and illumination show the effectiveness of the method.

This paper presents a new approach for the binarization of seriously degraded manuscript. We introduce a new technique based on a Markov random field (MRF) model of the document. Depending on the available information, the model parameters (clique potentials) are learned from training data or computed using heuristics. The observation model is estimated thanks to an expectation maximization (EM) algorithm which extracts text and paperpsilas features. The performance of the proposition is evaluated on several types of degraded document images where considerable background noise or variation in contrast and illumination exist.