We report a new technique for building a wide-angle, lightweight, thin form factor, cost effective, easy to manufacture near-eye Head-Mounted Display (HMD) for virtual reality applications. Our approach adopts an aperture mask containing an array of pinholes and a screen as a source of imagery. We demonstrate proof-of-concept HMD prototypes with a binocular field of view (FOV) of 70◦ × 45◦, or total diagonal FOV of 83◦. This FOV should increase with the increasing display panel size.

We introduce a model to evaluate the segmentation quality of a color image. The model parameters were learned from a set of examples. To this aim, we first segmented a set of images using a traditional graph-cut algorithm, for different values of the scale parameter. A human observer classified these images into three classes: under-, well- and over-segmented. We used such classification to learn the parameters of the segmentation quality model, that was then employed to automatically and adaptively optimize the scale parameter of the graph-cut segmentation algorithm.

We describe a supervised learning procedure for estimating the relation between a set of local image features and the local optimal parameters of an adaptive bilateral filter. A set of two entropy-based features is used to represent the properties of the image at a local scale. Experimental results show that our entropy-based adaptive bilateral filter outperforms other extensions of the bilateral lter where parameter tuning is based on empirical rules.

An implementation of the incomplete-LU/Cholesky preconditioned block-iterative methods on the Graphics Processing Units (GPUs) using the CUDA parallel programming model is presented. In particular, we focus on the tradeoffs associated with the sparse matrix-vector multiplication with multiple vectors, sparse triangular solve with multiple right-hand-sides (rhs) as well as incomplete factorization with 0 fill-in. We use these building blocks to implement the block-CG and block-BiCGStab iterative methods for the symmetric positive definite (s.p.d.) and nonsymmetric linear systems, respectively.

We present a real-time system that renders antialiased hard shadows using irregular z-buffers (IZBs). For subpixel accuracy, we use 32 samples per pixel at roughly twice the cost of a single sample. Our system remains interactive on a variety of game assets and CAD models while running at 1080p and 2160p and imposes no constraints on light, camera or geometry, allowing fully dynamic scenes without precomputation. Unlike shadow maps we introduce no spatial or temporal aliasing, smoothly animating even subpixel shadows from grass or wires.

We present a library of efficient implementations of deep learning primitives. Deep learning workloads are computationally intensive, and optimizing their kernels is difficult and time-consuming. As parallel architectures evolve, kernels must be reoptimized, which makes maintaining codebases difficult over time. Similar issues have long been addressed in the HPC community by libraries such as the Basic Linear Algebra Subroutines (BLAS). However, there is no analogous library for deep learning.

We present Aggregate G-Buffer Anti-Aliasing (AGAA), a new technique for efficient anti-aliased deferred rendering of complex geometry using modern graphics hardware. In geometrically complex situations, where many surfaces intersect a pixel, current rendering systems shade each contributing surface at least once per pixel. As the sample density and geometric complexity increase, the shading cost becomes prohibitive for real-time rendering. Under deferred shading, so does the required framebuffer memory.

Obtaining a good baseline between different video frames is one of the key elements in vision-based monocular SLAM systems. However, if the video frames contain only a few 2D feature correspondences with a good baseline, or the camera only rotates without sufficient translation in the beginning, tracking and mapping becomes unstable. We introduce a real-time visual SLAM system that incrementally tracks individual 2D features, and estimates camera pose by using matched 2D features, regardless of the length of the baseline.

Conventional pipelines for capturing, displaying, and storing images are usually defined as a series of cascaded modules, each responsible for addressing a particular problem. While this divide-and-conquer approach offers many benefits, it also introduces a cumulative error, as each step in the pipeline only considers the output of the previous step, not the original sensor data.

We present a method for video stabilization and rolling-shutter correction for videos captured on mobile devices. The method uses the data from an on-board gyroscope to track the camera's angular velocity, and can run in real time within the camera capture pipeline.