Animatable 3D Model Generation from 2D Monocular Visual Data

Detection of objects precisely in video surveillance system is vital for many applications such as abnormal activity detection, congestion analysis, person identification, military fields. For detection of objects, background subtraction and temporal detection methods are used. In background subtraction method moving and static objects which enters into a camera viewing region has been detected. In this method reference frame will be captured during camera initialization. It uses a reference frame as background, detects present frame and takes the difference between two frames. In this paper objects are detected in multiple regions separately from single camera instead of using multiple cameras to monitor different objects in different regions. And also user will have control to select required single or multiple regions to detect the object. This will reduces the overall cost of the system and increases the processing speed of the system if we choose a particular region. Accuracy of the object detection in different regions increases since it allows using different parameters like thresholding separately. So that it monitors different activities at the same time. Also detection of objects based on comparison, object of interest from stored data base is presented using single static camera view with good real time performance.

Camera auto-calibration using a sequence of 2D images with small rotations

We present a system and a set of tools for producing special effects in film/video making by applying image processing and human centered computing techniques. A combination of shot detection, object segmentation, background generation, and image warping techniques are used. The user selects the image frame or the object of interest, and the image warp transformation to be used from the GUI. Wrapping can be performed either on a whole image sequence or on an object of interest in the sequence. In the latter, the object is first segmented and its motion tracked. Object segmentation is achieved either by snakes or graph cuts. Steerable Pyramid background generation is then used to fill in the portion cut from the foreground.

This paper addresses the problem of motion estimation and reconstruction of 3D models from profiles of an object rotating on a turntable, obtained from a single camera. Its main contribution is the development of a practical and accurate technique for solving this problem from profiles alone, which is, for the first time, precise enough to allow the reconstruction of the object. No correspondence between points or lines are necessary, although the method proposed can be equally used when these features are available, without any further adaptation. Symmetry properties of the surface of revolution swept out by the rotating object are exploited to obtain the image of the rotation axis and the homography relating epipolar lines, in a robust and elegant way. These, together with geometric constraints for images of rotating objects, are then used to obtain first the image of the horizon, which is the projection of the plane that contains the camera centres, and then the epipoles, thus fully determining the epipolar geometry of the sequence of images. The estimation of the epipolar geometry by this sequential approach (image of rotation axis - homography - image of the horizon - epipoles) avoids many of the problems usually found in other algorithms for motion recovery from profiles. In particular, the search for the epipoles, by far the most critical step, is carried out as a simple one-dimensional optimisation problem. The initialisation of the parameters is trivial and completely automatic for all stages of the algorithm. After the estimation of the epipolar geometry, the Euclidean motion is recovered using the fixed intrinsic parameters of the camera, obtained either from a calibration grid or from self-calibration techniques. Finally, the spinning object is reconstructed from its profiles, using the motion estimated in the previous stage. Results from real data are presented, demonstrating the efficiency and usefulness of the proposed methods.

In recent years, many glaciers all over the world have been distinctly retreating and thinning. One of the consequences of this is the increase of so called glacier lake outburst flood events (GLOFs). The mechanisms ruling such GLOF events are still not yet fully understood by glaciologists. Thus, there is a demand for data and measurements that can help to understand and model the phenomena. Thereby, a main issue is to obtain information about the location and formation of subglacial channels through which some lakes, dammed by a glacier, start to drain. The paper will show how photogrammetric image sequence analysis can be used to collect such data. <br><br> For the purpose of detecting a subglacial tunnel, a camera has been installed in a pilot study to observe the area of the Colonia Glacier (Northern Patagonian Ice Field) where it dams the Lake Cachet II. To verify the hypothesis, that the course of the subglacial tunnel is indicated by irregular surface motion patterns during its collapse, the camera acquired image sequences of the glacier surface during several GLOF events. Applying tracking techniques to these image sequences, surface feature motion trajectories could be obtained for a dense raster of glacier points. Since only a single camera has been used for image sequence acquisition, depth information is required to scale the trajectories. Thus, for scaling and georeferencing of the measurements a GPS-supported photogrammetric network has been measured. <br><br> The obtained motion fields of the Colonia Glacier deliver information about the glacier’s behaviour before during and after a GLOF event. If the daily vertical glacier motion of the glacier is integrated over a period of several days and projected into a satellite image, the location and shape of the drainage channel underneath the glacier becomes visible. The high temporal resolution of the motion fields may also allows for an analysis of the tunnels dynamic in comparison to the changing water level of the lake.

A system for high speed capture of image sequences and real- time filed flattening has been designed and implemented. This system is useful for applications where the objects of interest in a scene are moving or changing, since it provides a means for obtaining the images as the objects move, as well as compensating for lighting variations. The system is implemented on a Datacube MV200 hosted by a SUN Sparc 5 workstation. The system has been successfully used to capture image sequences as part of a system to develop algorithms for dynamic face recognition.

Complex mission specifications can be often specified through temporal logics, such as Linear Temporal Logic and its syntactically co-safe fragment, scLTL. Finding trajectories that satisfy such specifications becomes hard if the robot is to fulfil the mission in an initially unknown environment, where neither locations of regions or objects of interest in the environment nor the obstacle space are known a priori. We propose an algorithm that, while exploring the environment, learns important semantic dependencies in the form of a semantic abstraction, and uses it to bias the growth of an Rapidly-exploring random graph towards faster mission completion. Our approach leads to finding trajectories that are much shorter than those found by the sequential approach, which first explores and then plans. Simulations comparing our solution to the sequential approach, carried out in 100 randomized office-like environments, show more than 50% reduction in the trajectory length.

This paper addresses the generation and visualization of noisy 3D point clouds. The goal is to extract 3D information from image sequences in real-time and to transform the resulting noisy point clouds into a representation with dense surfaces and an intuitive character. Therefore a bridge is built from image sequence analysis to computer graphics. A point cloud is calculated from a single camera by using state of the art algorithms for egomotion estimation and dense 3D reconstruction. Due to inaccuracies in the measurement process and hence of the resulting depth maps, a degree of uncertainty of the 3D data is defined. This additional information can be used to filter the point cloud in space and time. We present an approach to filter and polygonize weighted point clouds and visualize it by mapping image texture on the resulting surface.

Estimating the relative 3D shape of the object is an important research topic in the area of computer vision, and is being used in a wide range of applications such as robot vision, computer games, animations, broadcast, and many more. Several active and passive methods have been proposed for recovering 3-D shape of the objects from their 2-D images. Passive methods like Shape from Focus (SFF), Shape from Defocus (SFD), Shape from Shading (SFS) etc are cheap and more effective, requiring stack of images by a single camera. In this paper, we have developed a simple and fast algorithm for SFF to calculate depth. The pixel intensities in the image sequences are modified by subtracting the maximum of first or last frame in the image sequence. A low pass filter is applied on these modified values to eliminate the noise. The proposed algorithm is fast and precise, as compared to earlier SFF methods.

This paper presents a method of computing camera positions from a sequence of overlapping images obtained from a binocular/trinocular camera head. First, we find matching features among the images at each camera head position. Because the individual cameras are calibrated we can directly compute the 3D coordinates of these features using triangulation. Then we find matching features across adjacent images in the camera sequence. We compute the fundamental matrix between image pairs, and then the trilinear tensor between image triplets. The corresponding features that support the overlapping trilinear tensors are very reliable. Some of these matching features across the image sequence are also matching features among the images at each camera head location. This creates a potential set of matching 3D features between the adjacent images in the sequence. We compute the transformation between the camera positions in the image sequence using these matching 3D feature co-ordinates. Using multiple cameras has a number of advantages over computing the camera positions from a single camera. We directly obtain a Euclidean reconstruction of the camera path, we can reliably process very small motions, and there are no motion degeneracies.

Image sequence based automatic multi-camera system calibration techniques

A novel vision system is proven at a conceptual level to help unmanned remotely operated vehicles (ROVs) interpret underwater oceanic scenes and clarify noisy image sequences. The images contain objects of interest (metal cylinders of a known oil rig structure) and background (water). After contrast stretching for enhancement, images are segmented using a quickly convergent method based on Markov random fields (MRFs). This iterated conditional mode MRF uses deterministic relaxation to rapidly converge. Cylinders are analysed to determine the camera's viewing direction, range and twist off the vertical. The camera's position is then calculated given knowledge of the node in view. Successive viewpoints from a sequence of images are fed through a Kalman filter to predict the next viewpoint. Placing this in a 3D computer model of the structure allows a 2D predicted image to be projected. This is combined with the next acquired image to improve object recognition by the MRF segmentation method. This novel predictive feedback method shows better resilience to noise, matching noisy images to the clearer model that would otherwise go unrecognised.

This paper presents a real-time system for pedestrian tracking in sequences of grayscale images acquired by a stationary camera. The objective is to integrate this system with a traffic control application such as a pedestrian control scheme at intersections. The proposed approach can also be used to detect and track humans in front of vehicles. Furthermore, the proposed schemes can be employed for the detection of several diverse traffic objects of interest (vehicles, bicycles, etc.) The system outputs the spatio-temporal coordinates of each pedestrian during the period the pedestrian is in the scene. Processing is done at three levels: raw images, blobs, and pedestrians. Blob tracking is modeled as a graph optimization problem. Pedestrians are modeled as rectangular patches with a certain dynamic behavior. Kalman filtering is used to estimate pedestrian parameters. The system was implemented on a Datacube MaxVideo 20 equipped with a Datacube Max860 and was able to achieve a peak performance of over 30 frames per second. Experimental results based on indoor and outdoor scenes demonstrated the system s robustness under many difficult situations such as partial or full occlusions of pedestrians.

We present an image database system with the capability to locate specified object-level merge and separation events in a sequence of time-lapse images. Specifically, the objects of interest are live cells in phase contrast images acquired by scanning cytometry. The system is named TERSIS and it resides on a workstation accessing time lapse images on CD- ROM. The cell objects are segmented and the resulting data are processed to extract a time series and its time derivative series for each spatial feature. Cell objects are tracked through the image sequence by applying similarity metrics to the cell object feature vectors, and cell merge and separation events are located using global image statistics. Multiple hypotheses are generated and scored to determine participating cell objects in merge/separation events. The cell association and time-varying spatial data re stored in a database. A graphical suer interface provides the user with tools to specify queries for specific cellular states and events for recall and display. Primary limitation include the need for an automatic front-end segmenter and increased cell tracking volume. The design of this system is extensible to other object types and forms of sequential image input, including video.

The background subtraction is a widely used approach for Detecting Moving Objects (DMO) by a static camera using a simple algorithm; however, it is very sensitive to the local gradual changes of illumination, shadows, non-rigid moving objects and partial or full target occlusion. In order to overcome these issues, and to bring more potential to the solution, we propose an effective method for object detection based on active contour with camera motion compensation. The principle of active contours is to evolve an initial curve towards the object of interest that corresponds to the boundaries of the moving objects. Once the object has been detected, then it can be tracked by the Kalman filter. This latter requires many preassumptions about models and noise characteristics. As an alternative, a new method based on the Smooth Variable Structure Filter (SVSF) is implemented. The SVSF was introduced in an effort to provide a more robust estimation strategy. Detection and tracking algorithms require stable images to recognize a real moving target position. Therefore, the captured image from cameras which are placed on the moving platform, have undesired jitters, shakes and blurs. Video stabilization becomes an indispensable technique which is focusing on removing unnecessary camera vibrations from image sequences using a homographic matrix by extracting features using FAST corner detector and FREAK feature descriptor. The proposed algorithm is validated in real-world and the obtained results confirm the efficiency and robustness of our approach.