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.
To robustly estimate the pose, classical methods assume some geometrical and temporal assumptions (SfM: Structure from Motion, SLAM: Simultaneous Localization and mapping). These approaches take a pair of images as input and establish correspondences based on global strategy (using the whole image information) or sparse strategy (using key-points features). These correspondences allow solving a set...
Autonomous Underwater Vehicle (AUV) has limited energy capacity due to it being an embedded system. To overcome this limitation, the AUV can home into a docking station to recharge its battery. Several research has been conducted on the docking of AUV using vision. In some literatures, docking would fail if the target placed at the docking station is missing or disoriented from the camera view. This...
The knowledge of driver distraction will be important for self driving cars in the near future to determine the handoff time to the driver. Driver's gaze direction has been previously shown as an important cue in understanding distraction. While there has been a significant improvement in personalized driver gaze zone estimation systems, a generalized gaze zone estimation system which is invariant...
This paper presents an approach to detect and recognize actions of interest in real-time from a continuous stream of data that are captured simultaneously from a Kinect depth camera and a wearable inertial sensor. Actions of interest are considered to appear continuously and in a random order among actions of non-interest. Skeleton depth images are first used to separate actions of interest from actions...
In recent years, designing and testing video anomaly detection methods have focused on synthetic or unrealistic sequences. This has mainly four drawbacks: 1) events are controlled and predictable because they are usually performed by actors; 2) environmental conditions, e.g. camera motion and illumination, are usually ideal thus realistic conditions are not well reflected; 3) events are usually short...
Our objective is to count objects using a single frame from a surveillance camera. We focus on the area where individual object detectors fail, mostly due to clutter, occlusion, or variations in scene due to perspective change. For tackling the counting problem, first the object density is estimated by using ridge regression. Object counts are then estimated by integrating the density over the region...
The task of person re-identification (re-id) is to match images of people observed in different camera views. Recent researches mainly focus on feature representation and metric learning. Many global metric learning approaches have achieved good performance. Since comparing all of the samples with a single global metric is inappropriate to handle heterogeneous data, some local metric learning approaches...
A novel single hand Sign Language Recognition (SLR) method utilizing Weightless Neural Network (WNN) known as a RAMnet or n-tuple network is evaluated as a memory efficient technique. In contrast with standard multilayer perceptron neural network (MLPNN), the RAMnet does not require long iteration of presentation of training data in the training phase i.e. long training time, weights adjustments and...
When processing video, it is normally assumed that cameras are vertically oriented such that people appear upright, which helps simplify subsequent processing such as person detection. In real situations, due to the need to provide maximum coverage of the viewing space, cameras are usually placed with arbitrary orientations so the apparent vertical axis of the videos captured may not correspond to...
Domain adaptation (DA) algorithms address the problem of distribution shift between training and testing data. Recent approaches transform data into a shared subspace by minimizing the shift between their marginal distributions. We propose a method to learn a common subspace that will leverage the class conditional distributions of training samples along with reducing the marginal distribution shift...
In this paper, we demonstrate how automatic grasp selection can be achieved by placing a camera in the palm of a prosthetic hand and training a convolutional neural network on images of objects with corresponding grasp labels. Our labeled dataset is built from common graspable objects curated from the ImageNet dataset and from images captured from our own camera that is placed in the hand. We achieve...
This paper, for the first time, introduces a multiple-class boosting scheme (MBS) to combine depth motion maps (DMMs) and completed local binary patterns (CLBP) for action recognition. DMMs derive from projecting depth frames onto three orthogonal Cartesian planes (front, side and top) and characterize the motion energy of an action, on which the CLBP features are further extracted. And then a new...
A dry swimming machine is a machine that allows swim training or exercise on dry land. Few dry swimming machines are commercially available for specialized swim training and stroke style evaluation. These machines are usually expensive and not designed or made suitable for physical exercise. Most of these devices are sensor based. Few high-end swimming machines have been developed and customized for...
In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based...
In this paper, we explore a new algorithm to detect people with thermal cameras based on the standard Implicit Shape Model (ISM) technique. Our approach starts with the ISM to define the proposed centers of people locations. Then we utilize a novel method to detect people based on the density of the concentrated proposed centers by using an auto generated threshold mechanism. Our method is easy to...
Human tracking across multiple cameras is highly demanded for large scale video surveillance. To successfully track human across multiple uncalibrated cameras that have no overlapping field of views, a system to train more reliable camera link models is proposed in this paper. We employ a novel approach of combining multiple camera links and building bidirectional transition time distribution in the...
Person re-identification is an important problem of matching persons across non-overlapping camera views. However, the re-identification is still far from achieving reliable matching. First, many existing approaches are wholebody- based matching, and how body parts could affect and assist the matching is still not clearly known. Second, the learned similarity measurement/metric is equally used for...
The use of an artificial replica of a biometric characteristic in an attempt to circumvent a system is an example of a biometric presentation attack. Liveness detection is one of the proposed countermeasures, and has been widely implemented in fingerprint and iris recognition systems in recent years to reduce the consequences of spoof attacks. The goal for the Liveness Detection (LivDet) competitions...
Various hand-crafted features and metric learning methods prevail in the field of person re-identification. Compared to these methods, this paper proposes a more general way that can learn a similarity metric from image pixels directly. By using a "siamese" deep neural network, the proposed method can jointly learn the color feature, texture feature and metric in a unified framework. The...
At the current rate of technological advancement and social acceptance thereof, it will not be long before wearable devices will be common that constantly record the field of view of the user. We introduce a new database of image sequences, taken with a first person view camera, of realistic, everyday scenes. As a distinguishing feature, we manually transcribed the scene text of each image. This way,...
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.