Extruded shapes and patterns are widely used by graphic designers to increase the visual richness and realism of 3D models and virtual environments in general. However, the traditional approach that uses polygonal mesh is inappropriate for real-time rendering, in which, just a limited number of graphic primitives can be processed. Hence the interest of using per-pixel approaches. In this paper,...

Pixel level snakes (PLS) are a cellular active contour technique inspired on the energy-based deformable models. They are based on a pixel-level discretization of the contours and a massively parallel computation on every contour cell which lead to a high speed processing without penalizing the efficiency of the contour location. PLS can be fully implemented in current focal plane processor arr...

We have developed a method for the linear reconstruction of an image from undersampled, dithered data. The algorithm, known as Variable-Pixel Linear Reconstruction (or informally as “drizzling”), preserves photometry and resolution, can weight input images according to the statistical significance of each pixel, and removes the effects of geometric distortion both on image shape and photometry....

Manual generation of training examples for supervised learning is an expensive process. One way to reduce this cost is to produce training instances that are highly informative. To this end, it would be beneficial to produce training instances interactively. Rather than provide a supervised learning algorithm with one complete set of training examples before learning commences, it would be bett...

Bad pixels are generally treated as a loss of useable area and then excluded from averaged performance metrics. The definition and detection of “bad pixels” or “cosmetic defects” are seldom discussed, perhaps because they are considered self-evident or of minor consequence for any scientific grade detector, however the ramifications can be more serious than generally appreciated. While the defi...

Modeling the distribution of natural images is a landmark problem in unsupervised learning. This task requires an image model that is at once expressive, tractable and scalable. We present a deep neural network that sequentially predicts the pixels in an image along the two spatial dimensions. Our method models the discrete probability of the raw pixel values and encodes the complete set of dep...

Deconvolutional layers have been widely used in a variety of deep models for up-sampling, including encoder-decoder networks for semantic segmentation and deep generative models for unsupervised learning. One of the key limitations of deconvolutional operations is that they result in the so-called checkerboard problem. This is caused by the fact that no direct relationship exists among adjacent...

We describe a vertex and/or tracking detector designed to operate in an environment in which there is a large density of background hits.

We describe a method to perform per pixel divisions using graph ics hardware with a simple iterative algorithm In the worst case the number of rendering passes is for approximate divisions and for accurate divisions The algorithm can be used for D vec tor normalization evaluating complex lighting models and image reconstruction Introduction and Previous Work Graphics hardware can act as a power...

A new type of image sensor is emerging from the most advanced image sensor R&D laboratories in Japan. This sensor, termed an active pixel sensor (APS), may one day supplant charge-coupled devices (CCDs) in many applications. The charge-coupled device relies on the shifting of charge to read out the image. The active pixel sensor acts like a random access memory (RAM) device wherein each pixel c...