Jirí Havel

Yet Faster Ray-Triangle Intersection (Using SSE4)

Ray-triangle intersection is an important algorithm, not only in the field of realistic rendering (based on ray tracing) but also in physics simulation, collision detection, modeling, etc. Obviously, the speed of this well-defined algorithms implementations is important because calls to such a routine are numerous in rendering and simulation applications. Contemporary fast intersection algorithms, which use SIMD instructions, focus on the intersection of ray packets against triangles. For intersection between single rays and triangles, operations such as horizontal addition or dot product are required. The SSE4 instruction set adds the dot product instruction which can be used for this purpose. This paper presents a new modification of the fast ray-triangle intersection algorithms commonly used, which-when implemented on SSE4-outperforms the current state-of-the-art algorithms. It also allows both a single ray and ray packet intersection calculation with the same precomputed data. The speed gain measurements are described and discussed in the paper.

Review of Hough Transform for Line Detection

This chapter describes the basics of the Hough transform (HT). The terminology to be used in this text is defined in Sects. 2.1 and 2.2. The relationship of the HT (for lines) and the Radon and Fourier transforms is sketched out in Sect. 2.3. Section 2.4 reviews the most common existing line parameterizations used for line detection by the HT and gives a quick comparison of the important ones.

Nonnegative Tensor Factorization Accelerated Using GPGPU

This article presents an optimized algorithm for Nonnegative Tensor Factorization (NTF), implemented in the CUDA (Compute Uniform Device Architecture) framework, that runs on contemporary graphics processors and exploits their massive parallelism. The NTF implementation is primarily targeted for analysis of high-dimensional spectral images, including dimensionality reduction, feature extraction, and other tasks related to spectral imaging; however, the algorithm and its implementation are not limited to spectral imaging. The speedups measured on real spectral images are around 60 - 100× compared to a traditional C implementation compiled with an optimizing compiler. Since common problems in the field of spectral imaging may take hours on a state-of-the-art CPU, the speedup achieved using a graphics card is attractive. The implementation is publicly available in the form of a dynamically linked library, including an interface to MATLAB, and thus may be of help to researchers and engineers using NTF on large problems.

Uniform Marker Fields: Camera localization by orientable De Bruijn tori

In various applications, a wider area needs to be covered by fiduciary markers but a large marker cannot be used because only a fraction of the area is to be viewed by the camera. Such an area can be covered by a number of small markers with unique identifiers. However, with the camera freely moving in the scene and with occluders present, it is difficult to ensure that at least one of the individual markers is completely visible, unless the markers are small and numerous. In that case, the markers are not recognizable from larger distances. In this paper we introduce the concept of Marker Fields which overcome this limitation. The Marker Field covers a large-scale planar (or non-planar) area and it is composed of mutually overlapping partial markers. We propose a particular arrangement of the Marker Field: a Uniform Checker-Board Marker Field, which is a black- and-white checkerboard whose square modules are defined by aperiodic 4-orientable binary n2-window arrays (De Bruijn tori). We propose a genetic algorithm for construction of 4-orientable n2window arrays. We used a supercomputer to synthesize large 4-orientable 42window arrays and offer them publicly for downloading. We prototyped an algorithm for detection of the checkerboard marker fields and measured its performance. When processing input video from a cellphone camera, the algorithm visits only about 5 % of image pixels for reliable detection and the processing time is about 1 ms on a mid-range PC processor. The Uniform Marker Field increases freedom of camera movement, especially with occluders present in the scene. The detection algorithm is efficient and real-time marker field detection will be feasible on ultramobile devices.

Fractal marker fields: No more scale limitations for fiduciary markers

One limitation of existing fiduciary markers is that the camera motion is tightly limited: the marker (one of the markers) must be visible and it must be observed at a proper scale. This paper introduces a fractal structure of markers similar to matrix codes (such as QR-code or Data Matrix): the Fractal Marker Field. The FMF allows for embedding markers of a virtually unlimited number of scales. At the same time, for each of the scales it guarantees a constant density of markers at that scale over the whole marker fields surface. The Fractal Marker Field can provide unprecedented freedom of motion to camera-based augmented reality applications.

Low-Level Image Features for Real-Time Object Detection

Object detection in still images and in video sequences has a wide range of applications and while it is a very costly task from the computational resources point of view, very high demand exists for efficient object detection methods and implementations. One of the frequently used techniques of fast object detection is usage of classifiers to scan the image and attempt classification of every potential object position or even every potential position in the image being searched. The classifiers can be implemented as statistical classifiers based on supervised machine learning and can take as their input low-level features (sometimes called weak classifiers) extracted from the window being classified. In principle, such features can be immediately the image pixels, but by using more complex feature extractors, the classifiers can achieve better performance – both in the detection rate and in the speed. This chapter describes several image feature extractors used in real-time object detection and in detail discusses the novel features based on local ranks. The features have been designed so that they have equal descriptive and generalization power as their state-of-theart alternatives, but at the same time to be efficiently implementable in hardware. These features prove to be efficient, not only in the hardware implementations (tested in FPGA chips), but also when implemented using the SSE instruction set of the contemporary CPU’s and implemented in the graphics processors (GPU’s). The classification background and specification of requirements on the low-level image feature extractors is given in section 2. Formal definition of the features based on local ranks is given in section 3. The performance of the LRP image feature extractors was evaluated from the point of view of the classifier construction and the results are also given in section 3. Section 4 describes efficient implementations of the feature extractor on different hardware platforms, namely the SSE instruction set; FPGA (Field-Programmable Gate Arrays) defined in the hardware definition language VHDL; and GPU implementations, using both the shading language GLSL and the CUDA programming environment. Section 4 also contains performance evaluation of the different implementations of the low-level feature extractors. 6

Variants of the Hough Transform for Straight Line Detection

Various research groups invested effort to deal with computational complexity of the Hough transform and/or to improve the accuracy of the detection. Different methods focus on special data structures, non-uniform resolution of the accumulation array, special rules for picking points from the input image, or they employ specialized hardware units, etc.

Vanishing Points, Parallel Lines, Grids

This chapter will discuss the properties of perspectively projected rectangular grids and some observations fundamental for their detection. These grids are typically the edges in images containing checkerboards, QR codes, and similar patterns. It will also analyze the contents of the parameter space that results from the Hough transform of such patterns.

Efficient Implementation of PClines

This chapter discusses possibilities of implementation of the PClines-based detectors on graphics processors. Section 6.2 deals with an OpenGL implementation of the line detector [18]. The OpenGL implementation can be executed on various graphics devices, including advanced graphics chips for mobile phones.

PClines: Line Parameterization Based on Parallel Coordinates

Recently, parallel coordinates [39] were used for parameterizing lines for the Hough transform [19, 22]. This parameterization, referred to as the Parallel-Axis Transform (PAT) or PClines, is a point-to-line mapping. This chapter introduces the parameterization in detail. First, Sect. 4.1 reviews the basic information about the parallel coordinates needed for understanding the parameterization. Then, the parameterization is introduced in Sect. 4.2, and a brief analysis of its accuracy is given in Sect. 4.3