The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This paper introduces a novel system for sports motion analysis. The system utilizes wearable sensors and video cameras to detect a sports motion and to provide image sequences of the motion automatically. Deep neural networks were adopted for detection and explicit segmentation of a sports motion into submotions. The proposed system is evaluated by implementing a soccer kick analysis system as an...
This paper proposes a convolutional neural network architecture for blood vessel segmentation in retinal images. The network structure is designed on 7 layers using MatConvNet (three convolutional layers, two pooling layers, one dropout layer and a Softmax layer). The input data, selected from the DRIVE database, of the neural network is preprocessed in Matlab on Green channel. The retinal image was...
Face detection is an important step in face recognition. Ineffective algorithm used for face detection will have a negative impact on the performance of face recognition. Face detection technology is not only a key step in face recognition technology, but also is an independent widely-used technology. It is important to design a suitable feature for face detection technology. This paper proposes a...
This paper is devoted to investigation of features that will be the most appropriate for description of high resolution satellite imagery. We developed an image description model which is based on the distribution of image object classes. Proposed model could be used for image similarity estimation.
Melanoma skin cancer is on the rise globally due to increased ultraviolet radiation and even in darker skinned communities, new cases are being discovered. Like many cancers if detected early the chances of successful treatment and cure are high but if detected at a later stage the chances become low. In the application of Computer Aided Diagnosis systems for detection of melanoma, image pre-processing,...
Automatic liver segmentation from abdominal Computed Tomography (CT) is an important step for hepatic disease diagnosis. It is a challenging task owing to the similarity between liver and its adjacent organs and the low contrast of liver texture (e.g. tumors and blood veins). In this paper, we propose a cascaded structure to automatically segment liver in CT scans. First, we train a fully convolutional...
Traversable region estimation is the fundamental enabler in autonomous navigation. In this paper, we propose a traversable region segmentation algorithm using stereo vision. We address this problem mainly in road scenes for the goal of autonomous driving. Using only geometry information, our approach has the advantages of effectiveness and robustness. The proposed approach is based on a cascaded framework...
This study aims to meet requirement of rapid detection of the TBM (tunnel boring machine) tool's wear state. To this end, the local region-growing method based on normalized cross-correlation was proposed to detect the block slag on belt conveyor and the least-square method for ellipse fitting was employed to measure the size. Firstly, the monocular color image was decomposed into two gray-scale images...
Flexible printed circuit board (FPC) is a popular substrate for packaging integrated circuits (ICs). Detecting the circles rapidly on FPCs by using computer vision is very important to assess the quality of FPCs during its manufacturing. In this paper, a fast circle detection approach based on a threshold segmentation method and a validation check is proposed. In the algorithm, the image is firstly...
As the human eye on the image of different regions of the contrast sensitivity is different, it is particularly important to segment the image region more accurately in the image quality evaluation. Based on this, this paper presents a non-reference image region division method based on deep learning. Firstly, the Canny operator performs image edge detection at low threshold to obtain the strong edge...
PET imaging is increasingly used in determining functional tumor volumes for therapy response assessment and treatment planning. Accurate measurement of metabolically active tumors may improve optimal delivery of radiation treatment. However, volume measurement of small tumors (<2 cm) has been a much investigated subject due to Partial Volume Effect and limited resolution of the system. This research...
The Mapillary Vistas Dataset is a novel, large-scale street-level image dataset containing 25000 high-resolution images annotated into 66 object categories with additional, instance-specific labels for 37 classes. Annotation is performed in a dense and fine-grained style by using polygons for delineating individual objects. Our dataset is 5× larger than the total amount of fine annotations for Cityscapes...
Deep convolutional neural networks (CNNs) have been successfully applied to a wide variety of problems in computer vision, including salient object detection. To detect and segment salient objects accurately, it is necessary to extract and combine high-level semantic features with low-levelfine details simultaneously. This happens to be a challenge for CNNs as repeated subsampling operations such...
The interactive image segmentation model allows users to iteratively add new inputs for refinement until a satisfactory result is finally obtained. Therefore, an ideal interactive segmentation model should learn to capture the user's intention with minimal interaction. However, existing models fail to fully utilize the valuable user input information in the segmentation refinement process and thus...
We introduce Appearance-MAT (AMAT), a generalization of the medial axis transform for natural images, that is framed as a weighted geometric set cover problem. We make the following contributions: i) we extend previous medial point detection methods for color images, by associating each medial point with a local scale; ii) inspired by the invertibility property of the binary MAT, we also associate...
We present a benchmark suite for visual perception. The benchmark is based on more than 250K high-resolution video frames, all annotated with ground-truth data for both low-level and high-level vision tasks, including optical flow, semantic instance segmentation, object detection and tracking, object-level 3D scene layout, and visual odometry. Ground-truth data for all tasks is available for every...
RGBD semantic segmentation requires joint reasoning about 2D appearance and 3D geometric information. In this paper we propose a 3D graph neural network (3DGNN) that builds a k-nearest neighbor graph on top of 3D point cloud. Each node in the graph corresponds to a set of points and is associated with a hidden representation vector initialized with an appearance feature extracted by a unary CNN from...
In this paper, we approach the problem of segmentation-free query-by-string word spotting for handwritten documents. In other words, we use methods inspired from computer vision and machine learning to search for words in large collections of digitized manuscripts. In particular, we are interested in historical handwritten texts, which are often far more challenging than modern printed documents....
In this paper we are interested in the problem of image segmentation given natural language descriptions, i.e. referring expressions. Existing works tackle this problem by first modeling images and sentences independently and then segment images by combining these two types of representations. We argue that learning word-to-image interaction is more native in the sense of jointly modeling two modalities...
Humans take advantage of real world symmetries for various tasks, yet capturing their superb symmetry perception mechanism with a computational model remains elusive. Motivated by a new study demonstrating the extremely high inter-person accuracy of human perceived symmetries in the wild, we have constructed the first deeplearning neural network for reflection and rotation symmetry detection (Sym-NET),...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.