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.
Vision — based road lane detection and reconstruction is a very common interest in the field of computer vision (CV). It has numerous application ranging from autonomous vehicle to driver assist and support systems technology. These researches are always focusing on both accuracy and complexity of the system's output; however, none of these uses Macro Block (MB) method. This paper introduces the characteristics...
The algorithm of extracting pedestrian features based on Local Binary Pattern (LBP) has the problems of being unable to depict the human visual sensitivity. We proposed an Significant Local Binary Pattern (SLBP) which fused the characteristics of human visual pedestrian system. We extracted the significant factor based on Weber's law, and added the significant factor as a weight to the LBP eigenvalue...
In this work we show how sublabel-accurate multilabeling approaches [15, 18] can be derived by approximating a classical label-continuous convex relaxation of nonconvex free-discontinuity problems. This insight allows to extend these sublabel-accurate approaches from total variation to general convex and nonconvex regularizations. Furthermore, it leads to a systematic approach to the discretization...
In this paper, we propose a patched-based deep Boltzmann shape priors for visual tracking. The shape priors are generated from deep Boltzmann machine network. The network consists of three layers of hidden and visible units. The generated shapes not only maintain general shapes from a variety of poses, but also entail local modifications with high probability.
This target detection and tracking system is the basis for rescue robots to achieve their independent search and rescue operations. In order to improve their mobile performance and sensing capability, the Kinect camera is employed by rescue robots to obtain environmental visual. The AKAZE(Accelerated-KAZE) feature matching algorithm is adopted to achieve target detection in video frames, combining...
We consider the problem of finding consistent matches across multiple images. Current state-of-the-art solutions use constraints on cycles of matches together with convex optimization, leading to computationally intensive iterative algorithms. In this paper, we instead propose a clustering-based formulation: we first rigorously show its equivalence with traditional approaches, and then propose QuickMatch,...
This paper proposes an accurate and generalizable deep learning framework for iris recognition. The proposed framework is based on a fully convolutional network (FCN), which generates spatially corresponding iris feature descriptors. A specially designed Extended Triplet Loss (ETL) function is introduced to incorporate the bit-shifting and non-iris masking, which are found necessary for learning discriminative...
The rapid and irregular motion of semen cells makes the counting process of semen difficult in the visual assessment. Therefore, computer based techniques are necessary to evaluate the tests with more accurately. In this paper, an alternative way to the visual assessment technique in spermiogram tests is presented. Analyses are performed on the recorded microscope video images by computer, automatically...
Technique of comparing pedestrian images observed by different cameras to determine whether they are the same person is important in the surveillance system. This technique is called Person re-identification. Most of Person reidentification is underway assuming that occlusion does not occur. However, since occlusion occurs frequently in the surveillance system and affects accuracy, it is necessary...
This paper considers the problem of recovering either a low rank matrix or a sparse vector from observations of linear combinations of the vector or matrix elements. Recent methods replace the non-convex regularization with ℓ1 or nuclear norm relaxations. It is well known that this approach recovers near optimal solutions if a so called restricted isometry property (RIP) holds. On the other hand it...
Faster R-CNN (R corresponds to “Region”) which combined the RPN network and the Fast R-CNN network is one of the best ways to object detection of R-CNN series based on deep learning. The proposal obtained by RPN is directly connected to the ROI Pooling layer, which is a framework for CNN to achieve end-to-end object detection. The feasibility of Faster R-CNN implementation of ResNet101 network and...
We propose a simple, yet powerful regularization technique that can be used to significantly improve both the pairwise and triplet losses in learning local feature descriptors. The idea is that in order to fully utilize the expressive power of the descriptor space, good local feature descriptors should be sufficiently “spread-out” over the space. In this work, we propose a regularization term to maximize...
A tile-based semi-global matching (SGM) processor with lossless data compression is proposed. The 8×8 tile-base processing and the P2-less data compression can reduce the external memory access by 85% without any change in the processing result. In addition, the P2-less data compression can decrease on-chip SRAM size by 50%. Implemented in 65nm CMOS technology, the 6.3mm2 chip consumes 288mW and supports...
In this paper, a robust visual tracking system by utilizing the images acquired from a color camera and a thermal camera is proposed to track the target with real-time performance. The thermal camera, which can observe the heat originated from the target such as the human body or vehicle, can collaborate with the color camera to track the target in the cluttered environment or under occlusion. Unlike...
Real-time people counting from video records is a main building bloc for many applications in smart cities. In practice, this task usually encounters many problems, like the lack of real-time processing of the recorded videos or the occurrence of errors due to irrelevant people being counted. To overcome the above issues, we propose a novel real-time people counting approach dubbed YOLO-PC (YOLO based...
The rapid development of three-dimensional (3D) imaging techniques has significantly increased the demand for high resolution (HR) depth video and images. Significant pixel deficiencies and too much noise can be seen in depth images especially taken from Kinect cameras. For this reason, usability in several computer vision applications is restricted. In the acquisition of HR depth images, in traditional...
Multi-class multi-object tracking is an important problem for real-world applications like surveillance system, gesture recognition, and robot vision system. However, building a multi-class multi-object tracker that works in real-time is difficult due to low processing speed for detection, classification, and data association tasks. By using fast and reliable deep learning based algorithm YOLOv2 together...
This paper proposes a new stereo corresponding algorithm which uses local-based. The Sum of Absolute Differences (SAD) algorithm produces accurate results on the disparity map for the textured regions. However, this algorithm is sensitive to low texture areas and high noise on images with high different brightness and contrast of images. To get over these problems, the proposed algorithm utilizes...
Saliency detection in images attracts much research attention for its usage in numerous multimedia applications. In this paper, we propose a saliency detection method based on optimization for RGBD images. With RGBD images, our method utilizes the depth channel to enhance the identification of background and foreground regions. We firstly generate new depth image by using non-linear transformation...
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.