Modern GPUs accelerate computations and data movements of multi-dimensional tensors in hardware. However, expressing optimized tensor computations in software is extremely challenging even for experts. Languages like CUDA C++ are centered around flat buffers in one-dimensional memory and lack reasonable abstractions for multi-dimensional data and threads. Existing tensor IRs are not expressive enough to represent the complex data-to-thread mappings required by the GPU tensor instructions.

Array-based programming languages have shown significant promise for improving performance of column-based in-memory database systems, allowing elegant representation of query execution plans that are also amenable to standard compiler optimization techniques. Use of loop fusion, however, is not straightforward, due to the complexity of built-in functions for implementing complex database operators. In this work, we apply a compiler approach to optimize SQL query execution plans that are expressed in an array-based intermediate representation. We analyze this code to determine shape properties of the data being processed, and use a subsequent optimization phase to fuse multiple database operators into single, compound operations, reducing the need for separate computation and storage of intermediate values. Experimental results on a range of TPC-H queries show that our fusion technique is effective in generating efficient code, improving query time over a baseline system.

Relational database management systems (RDBMS) are operationally similar to a dynamic language processor. They take SQL queries as input, dynamically generate an optimized execution plan, and then execute it. In recent decades, the emergence of in-memory databases with columnar storage, which use array-like storage structures, has shifted the focus on optimizations from the traditional I/O bottleneck to CPU and memory. However, database research so far has primarily focused on CPU cache optimizations. The similarity in the computational characteristics of such database workloads and array programming language optimizations are largely unexplored. We believe that these database implementations can benefit from merging database optimizations with dynamic array-based programming language approaches. Therefore, in this paper, we propose a novel approach to optimize database query execution using a new array-based intermediate representation, HorseIR, that resides between database queries and compiled code. Furthermore, we provide a translator to generate HorseIR from database execution plans and a compiler that optimizes HorseIR and generates efficient code. We compare HorseIR with the MonetDB RDBMS, by testing standard SQL queries, and show how our approach and compiler optimizations improve the runtime of complex queries.

Recent advances in execution environments for JavaScript and WebAssembly that run on a broad range of devices, from workstations and mobile phones to IoT devices, provide new opportunities for portable and web-based numerical computing. Indeed, numerous numerical libraries and applications are emerging on the web, including Tensorflow.js, JSMapReduce, and the NLG Protein Viewer. This paper evaluates the current performance of numerical computing on the web, including both JavaScript and WebAssembly, over a wide range of devices from workstations to IoT devices. We developed a new benchmarking approach, which allowed us to perform centralized benchmarking, including benchmarking on mobile and IoT devices. Using this approach we performed four performance studies using the Ostrich benchmark suite, a collection of numerical programs representing the numerical dwarf categories identified by Colella. We studied the performance evolution of JavaScript, the relative performance of WebAssembly, the performance of server-side Node.js, and a comprehensive performance showdown for a wide range of devices.

A nonlinearity compensated smooth frame insertion (NCSFI) method is proposed to reduce motion-blur caused by hold-type display in LCD. Firstly, each original frame is duplicated into two frames. Then in the first frame, spatial high frequency is removed and in the second frame, the same amount of high frequency is enhanced. Finally, look-up table based nonlinearity compensation is applied to the two new frames to compensate the nonlinearity between gray level and panel luminance. Experiments on real-time display system show that the proposed NCSFI method can reduce motion-blur significantly without luminance distortion

A supervised video object segmentation algorithm using a small number of interactions is proposed. The proposed algorithm is composed of three steps: semi-automatic first frame segmentation; automatic object tracking; boundary refinement. Homogeneous region segmentation is performed before the user interaction for the first frame to minimize the amount of interaction. Then, a polygon with very few key nodes can be drawn conveniently to get the segmentation mask of the first frame. In the object region tracking, pixel-wise backward tracking is adopted. Finally, a method for mask refinement is proposed by considering similar pixels in the neighborhood of each pixel. Extensive experimental results show that the proposed algorithm is effective for video object segmentation semiautomatically.

An algorithm for region-based moving object segmentation is presented. The gray-scale image segmentation based on the mean shift algorithm (MSA) is first performed to segment each frame of a sequence into connective homogeneous regions. A method applying spatio-temporal continuity constraints of the motion vector image is then carried out to detect moving pixels robustly. Finally, each homogeneous region is labeled as either a moving-object region or a non-moving-object region according to the number of moving pixels it contains. Experimental results show that our algorithm is effective and robust in segmenting moving vehicles from noisy scenes.

Discusses mosaicking and warping for forward-motion (or forward-moving) images (FMI). In FMI, the field of view of one frame is wider than that of a later one; however, the resolution of the later frame is higher than that of previous one. Warp-function algorithms based respectively on a mesh warp and a triangle grid are presented, and a multi-resolution mosaic based on the wavelet transform is given. In-between images could be warped from the single mosaic image. An experiment with a real scene shows that the mesh warp algorithm is more time-consuming than that of triangle grid, but the image quality of the latter is better than that of the former. It also shows that the multi-resolution mosaic image is smooth and natural, eliminating the impact of the seams.