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.
We present a Bayesian framework for estimating 3D human pose and camera from a single RGB image. We develop a generative model where a 3D pose is rendered onto an image (via the camera), which then generates a detection probability map for each body part. We represent a human pose with a set of 3D cylinders in space, one for each body part, and we place kinematic and self-intersection priors on the...
This paper addresses the problem of selecting instances of a planar object in a video or from a set of images based on an evaluation of its “frontalness”. We introduce the idea of “evaluating the frontalness” by computing how close the object's surface normal aligns with the optical axis of a camera. The unique and novel aspect of our method is that unlike previous planar object pose estimation methods,...
This paper proposes a moving object detection algorithm for a monocular dashcam mounted on a vehicle. To deal with dynamic changes of the scene from the dashcam, we propose a new scheme inspired by human-attention inclination for change detection. Humans do not build a detailed visual representation and perceive a change of the scene based on the structure of an interesting region. In this perspective,...
Traditional gaze tracking systems rely on explicit infrared lights and high resolution cameras to achieve high performance and robustness. These systems, however, require complex setup and thus are restricted in lab research and hard to apply in practice. In this paper, we propose to perform gaze tracking with a consumer level depth sensor (Kinect). Leveraging on Kinect's capability to obtain 3D coordinates,...
In order to fast register a camera into a 3D scene model under the Manhattan-World assumption, a method of matching corresponding 2D and 3D lines based on vanishing point is proposed in this paper. Firstly, this method detects line segments and estimates three orthogonal vanishing points to determine the local length of camera and the matrix from world to camera space. Afterwards, one line is drawn...
This paper proposes the use of change detection in a multi-view object recognition system in order to improve its flexibility and effectiveness in dynamic environments. Multi-view recognition approaches are essential to overcome problems related to clutter, occlusion or camera noise, but the existing systems usually assume a static environment. The presence of dynamic objects raises another issue...
Deep learning is greatly successful when used for pedestrian detection. However, we find that this method is barely satisfactory for multi-scale detection. Meanwhile, various solutions such as multi-scale classifiers have been developed (based on traditional methods) to handle this situation. Considering this, we propose a scale-discriminative classifier layer (SDC) that contains numerous classifiers...
Urban scene modeling is a promising research subject, for it plays an important role in the areas of national defense, transportation, urban planning, and entertainment industries. Researchers have been looking forward to the breakthroughs in improving the level of photorealism and the performance. This paper discusses the implementation of the simulation of certain parts of city scene. We use the...
Current methodologies for the generation of 3D point cloud from real world scenes rely upon a set of 2D images capturing the scene from several points of view. Novel plenoptic cameras sample the light field crossing the main camera lens creating a light field image. The information available in a plenoptic image must be processed in order to render a view or create the depth map of the scene. This...
Automated hand detection is useful for applications requiring reliable hand posture and hand gesture processing. Such applications include human-computer/human-robot interfaces for rehabilitation, serious games, or non-invasive medical diagnosis. Hence, in this paper, we focus on the design and development of a robust and fast hand detection and tracking (HD) system. The design of our HD system involved...
The paper presents a WebGL model of Segway based laboratory device that is available to user via web browser. The user can interact with the model and simulate various initial conditions. In addition, its dynamical behavior can also be controlled using PID controller. The background of the application is driven by OpenModelica software environment.
This paper deals with an advanced approach for the verification of ADAS algorithms in complex automotive sensor networks. We introduce a real-time modular framework that can benchmark the performance of the ADAS algorithms at the electronic system level using a Hardware-in-the-Loop (HiL) co-simulation. The presented research provides a generic architecture of ADAS sensor error injection for robustness...
This paper introduces an approach to employ deep features for person re-identification. In contrast to existing works, we focus on using pre-trained deep models and their concept-based output to enhance attribute presentations of person images. There are two main contributions. First, we investigate recent state-of-the-art deep learning models for the task and provide a comprehensive evaluation. Second,...
In this paper, an efficient road detection algorithm is proposed. By exploiting the recent progress of fast stereo matching algorithms, the proposed road detection algorithm relies on the accurate semi-dense disparity map only. The RANSAC algorithm is used to compute road plane parameters on randomly sampled disparity patches. Unreliable patches are removed by introducing a road plane slope constraint,...
As a key parameter of agile satellite TDICCD camera, the integral time calculation accuracy affects the satellite imaging quality directly. The traditional integral time calculation model is unable to satisfy the accuracy requirement for attitude changes flexibility and integral time changes severely during the satellite imaging. This paper describes an integral time calculation model that improved...
In this paper, it is shown that the Enhanced Unified Camera Model, which is used to model fisheye cameras, projects straight lines into conic sections. Then a way to find equations of straight line projections is proposed. The method is applied to epipolar curves to adapt the Semi-Global Matching algorithm to fisheye stereo systems to compute a dense direct stereo correspondence without undistortion...
A method for processing plenoptic images and reconstructing 3D models of real objects is presented in this paper. The reconstruction of the object is made by using the capability of the plenoptic camera to perceive the depth. Through six shots, which shoot the object in a central perspective, it will be shown how it is possible to recreate the object in a digital 3D space, in its entirety, at low...
Taking digital photographs as a form of documentation while conducting field work is common practice. There is a wealth of digital photographs collected by foresters who are interested in tree architecture. These high-resolution photographs are two-dimensional representations of the tree structure. However, it is possible to exploit tools such as GIS (e.g., ESRI ArcGIS) and three-dimensional modelling...
Applying computer vision for control the school of fish robots (SFR) using swarm model is a new direction. In this paper, the computer vision was studied for control a SFR using swarm model. The computer vision will return the framework and an algorithm was used to take the real position for every member. In the swarm model, every member has to solve a problem and combination together to get the target...
Existing approaches to object detection address the generation of object hypotheses by extracting several cues in natural and automotive images, relying on objects with sufficiently high resolution. Very little to almost no approaches, however, address the generation of hypothesis of very small or distant objects in images such as on motorways. Here, we propose a simple yet effective approach to generating...
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.