In this paper, we propose a new method, video repairing, to robustly infer missing static background and moving fore-ground due to severe damage or occlusion from a video. To recover background pixels, we extend the i...
详细信息
In this paper, we propose a new method, video repairing, to robustly infer missing static background and moving fore-ground due to severe damage or occlusion from a video. To recover background pixels, we extend the image repairing method, where layer segmentation and homography blending are used to preserve temporal coherence and avoid flickering. By exploiting the constraint imposed by periodic motion and a subclass of camera and object motions, we adopt a two-phase approach to repair moving foreground pixels: In the sampling phase, motion data are sampled and regularized by 3D tensor voting to maintain temporal coherence and motion periodicity. In the alignment phase, missing moving foreground pixels are inferred by spatial and temporal alignment of the sampled motion data at multiple scales. We experimented our system with some difficult examples, where the camera can be stationary or in motion.
In this paper, we present a system for reconstructing photorealistic 3D head models from multi-view images. In the proposed system, we perform two-pass bundle adjustments to reconstruct a photo-realistic 3D head model...
详细信息
This paper introduces feature-based textures, a new image representation that combines features and samples for high-quality texture mapping. Features identify boundaries within an image where samples change discontin...
详细信息
In this paper we present an occlusion culling method that uses hardware-based depth queries on oriented bounding boxes to cull unseen geometric primitives efficiently. An out-of-core design enables this method to inte...
详细信息
In this paper we present an occlusion culling method that uses hardware-based depth queries on oriented bounding boxes to cull unseen geometric primitives efficiently. An out-of-core design enables this method to interactively display data sets that are too large to fit into main memory. During a preprocessing phase, a spatial subdivision (such as an octree or BSP tree) of a given data set is constructed where, for each node, an oriented bounding box containing mesh primitives is computed using principal component analysis (PCA). At runtime, the tree indicated by the spatial subdivision is traversed in front-to-back order, and only nodes that are determined to be visible, based on a hardware accelerated depth query, are rendered.
The increasing popularity of points as rendering primitives has led to a variety of different rendering algorithms, and in particular the different implementations compare like apples to oranges. In this paper we revi...
详细信息
Parametric snake models are deformable models very useful for boundary extraction and tracking. Despite their abilities, the non-invariance of the internal energy under affine transformations and the non-convexity of ...
详细信息
We present a new algorithm for computing a polygonal surface from a union of balls. The method computes and connects the singular points of a given union of balls in an efficient manner to approximate the boundary. Th...
详细信息
We present a new algorithm for computing a polygonal surface from a union of balls. The method computes and connects the singular points of a given union of balls in an efficient manner to approximate the boundary. The algorithm uses the dual shape of the balls to give the resulting surface the correct topology. Our method is simple and demonstrated to be robust
Most medical image processing techniques require the clinician to form a mental model of the anatomy being depicted. We present systems that allow a more direct visualization of anatomical shapes and measurement of pa...
详细信息
Geometric coordinates are an integral part of many data streams. Examples include sensor locations in environmental monitoring, vehicle locations in traffic monitoring or battlefield simulations, scientific measuremen...
详细信息
ISBN:
(纸本)9781581138580
Geometric coordinates are an integral part of many data streams. Examples include sensor locations in environmental monitoring, vehicle locations in traffic monitoring or battlefield simulations, scientific measurements of earth or atmospheric phenomena, etc. How can one summarize such data streams using limited storage so that many natural geometric queries can be answered faithfully? Some examples of such queries are: report the smallest convex region in which a chemical leak has been sensed, or track the diameter of the dataset. One can also pose queries over multiple streams: track the minimum distance between the convex hulls of two data streams;or report when datasets A and B are no longer linearly separable. In this paper, we propose an adaptive sampling scheme that gives provably optimal error bounds for extremal problems of this nature. All our results follow from a single technique for computing the approximate convex hull of a point stream in a single pass. Our main result is this: given a stream of two-dimensional points and an integer r, we can maintain an adaptive sample of at most 2r + 1 points such that the distance between the true convex hull and the convex hull of the sample points is O(D/r2), where D is the diameter of the sample set. With our sample convex hull, all the queries mentioned above can be answered in either O(log r) or O(r) time.
The problem of the parameterization of data points in NURBS curve/surface has been considered by amount of researchers. In this study, a new parameterization method considered as hybrid parameterization method is prop...
详细信息
The problem of the parameterization of data points in NURBS curve/surface has been considered by amount of researchers. In this study, a new parameterization method considered as hybrid parameterization method is proposed. Like universal method, this method uses the property of “ at each span index, there is exactly one maximum value of the rational B-spline basis function”. By taking the maximum rational B-spline basis functions as initial values, the centripetal method is applied to generate the parameter values of hybrid parameterization. This approach inherits the advantages of universal and centripetal methods such as allowing having multiple knots value. The proposed method gives better object shaped relevant to the other methods.
暂无评论