Computer Image Generation Research Articles

ОБОБЩЕННЫЙ АЛГОРИТМ РАСПРЕДЕЛЕНИЯ РЕСУРСОВ КОМПЬЮТЕРНОГО ГЕНЕРАТОРА ИЗОБРАЖЕНИЯ ПО 3D-МОДЕЛИ ТЕРРИТОРИИ МЕСТНОСТИ

The development of modern software systems that use computer graphics algorithms in real time to create 3D-models of the terrain expands the range of training tasks available to the Customer for training pilots on aviation simulators. Particular attention is paid to training pilots in the skills of solving navigation problems while flying over certain 3D terrain, using visual references in the form of three-dimensional models of reference objects. To achieve this goal, it is necessary to solve several important aspects: the selection of 3D-objects on the terrain as reference points, the development of an information resource algorithm and the optimal distribution of computer image generator resources throughout the 3D-model of the terrain to ensure the possibility of synthesizing a sufficient number of 3D-models reference objects in real time. Solutions are presented for modeling a terrain area of sufficient size for distributing information resources and training the pilot to solve navigation problems.

Программно-аппаратные системы машинного синтеза изображений для авиационных тренажеров

Options for the development of software and hardware systems (computer image generators) for machine synthesis in real time (with a cycle of 80-120 ms) of images of 3D-models of reference objects observed by the pilot during a flight on an aviation simulator through the cockpit glazing and on instruments simulating cabin equipment have been studied. Domestic and foreign computer image generators that implement the process of machine image synthesis are considered. It is shown that the technological process of synthesizing 2D-projections of 3D-models of reference objects in all flight simulator simulators is the same, despite the observation by the pilot of 3D-models of reference objects (with the possibility of professional eye training), or 2D-projections of 3D-models of reference objects with the ability to determine distances between the aircraft model and the selected 3D-model of reference objects using special marks.

Simulators of visual environment of aviation and space simulators

The development of synthesis systems for human 3D-models of reference objects allowing them to be used in simulators of visual environment of aviation and space simulators follows the way of development: 3D-devices of indication; computer image generators and databases. Negative conditions of the USSR collapse led to the fact that the development of domestic computer image generators was stopped. However, the accumulated experience of development and serial production of visual environment simulators makes it possible to define available trends of their further development. Modern achievements of science and technology allow not only to solve the problem of synthesis of 2D-projections of 3D-models on the screen plane connected to the computer image generator but also, depending on the Customer’s requirements, to synthesise for a person visually observed 3D-objects with the possibility to form separate components of his cognitive model, in which his experience of orientation in space is accumulated. For example, to orientate in space by selected reference objects with simultaneous training of the trainee’s eye-sight, or to orientate in space by observing 2D-projections of 3D-models of reference objects additionally using information from simulators of cabin equipment devices. The article gives an overview of the main parts of visual simulators of aviation and space simulators.

EVALUATING DEEP SEMI-SUPERVISED LEARNING METHODS FOR COMPUTER VISION APPLICATIONS

Deep semi-supervised learning (SSL) have been significantly investigated in the past few years due to its broad spectrum of theory, algorithms, and applications. The extensive use of the SSL methods is dominant in the field of computer vision, for example, image classification, human activity recognition, object detection, scene segmentation, and image generation. In spite of the significant success achieved in these domains, critically analyzing SSL methods on benchmark datasets still presents important challenges. In the literature, very limited reviews and surveys are available. In this paper, we present short but focused review about the most significant SSL methods. We analyze the basic theory of SSL and the differences among various SSL methods. Then, we present experimental analysis to compare these SSL methods using standard datasets. We also provide an insight into the challenges of the SSL methods.

Computer image generation for job simulation: An effective approach to occupational Risk Analysis

The paper deals with the general approach and features of the computer image generation for job simulation (CIGJS) method, as specially developed to support the Job Safety Analysis (JSA) technique. Starting from some general considerations on the occupational safety and health statistical data collection in Italy, when compared to US methodology, the paper provides an overview of the CIGJS method, developed to improve the effectiveness and usability of the traditional JSA. A case study is discussed in order to highlight how CIGJS makes the use of the JSA technique possible also at the design stage (this being fundamental according to a Prevention through Design approach), and allows to control the effectiveness of the adopted countermeasures. The application of the method complies with the basic European Directives requirements both on workers’ safety and health and on machinery analysis. Furthermore, CIGJS may substantially aid workers’ education and training, to the point of being a possible simulation-training device.

Aliasing and scintillation reduction in real-time computer graphics

Aliasing is unavoidable in real-time computer image generation due to the sampling processes occurring within the graphics hardware. In particular, aliasing produces scintillation effects and significant radiometric inaccuracy when targets are rendered at long range. This problem was alleviated some years ago by the development of a zoom anti-aliasing (ZAA) technique within infrared missile seeker hardware-in-the-loop simulations. An alternative ZAA technique based on extensive use of available graphics hardware functions is described here and compared to the original technique.

Visualization in Transportation: Current Practice and Future Directions

The use of computer image generation technologies within the transportation field to aid both the design and the user communities to visualize better the essential elements of a proposed facility’s structure as well as its operation is addressed. The primary focus is on current and future applications of these technologies and their integration within the overall notion of context-sensitive design and public involvement. The discussion suggests that a movement away from the visualization of what something looks like to the visualization of how something works or operates is being encountered, a trend that forces the integration of modeling and simulation. The discussion also suggests the importance of applications that use visualization to help bridge the gap between notions of design and notions of value. Such applications also reflect an increased integration of design-oriented applications and planning-oriented applications, such as those that incorporate the use of spatially referenced data systems such as geographic information systems.

A spectrally based framework for realistic image synthesis

Realistic image synthesis provides principles and techniques for creating realistic imagery based on models of real-world objects and behaviors. It has widespread applications in 3D design, computer animation, and scientific visualization. While it is common to describe light and objects in terms of colors, this approach is not sufficiently accurate and cannot render many spectral phenomena such as interference and diffraction. Many researchers have explored spectral rendering and proposed several methods for spectral representation, but none satisfy all representation criteria such as accuracy, compactness and efficiency. Furthermore, previous studies have focused on distinct behaviors of natural phenomena but few on their commonality and generality, and it is difficult to combine existing algorithms to simulate complex processes. This thesis proposes solutions to these problems within a spectrum-based rendering framework. The pipeline begins by loading spectra from a database to specify light sources and objects, then generates a spectral image based on local and global illumination models, projects the spectral image into a CIE image, and finally converts the CIE image into an RGB image for display or a CIELab image for evaluation. In spite of omitting the light phase information, it is shown that this approach suffices to generate all optical effects important for realistic image synthesis. As components of the new framework, a heuristic method is proposed for deriving spectra from colors and an error metric is provided for evaluating synthesized images. The new spectral representation proposed in this thesis is called the composite model. Its key point is to decompose any spectrum into a smooth background and a collection of spikes. The smooth part can be represented by Fourier coefficients and a spike by its location and height. A re-sampling technique is proposed to improve performance. Based upon the characteristics of human perception, the sufficiency of a low-dimensional representation is shown analytically. This model improves upon existing methods with aspect to accuracy, compactness and efficiency, and offers an effective vehicle for rendering optical effects involving spiky spectra. Using the proposed framework and composite spectral model, this thesis develops new illumination models for rendering a number of optical effects including dispersion, interference, diffraction, fluorescence, and volume absorption. The rendered images closely correspond to their real-world counterparts. Overall, this thesis improves realistic image synthesis by expanding its rendering capabilities. It may serve as a basis for a more sophisticated rendering environment for high-quality computer image generation.

The Effect of Visual System Time Delay on Helicopter Control

There is interest in the development of synthetic visual systems to improve the capability of aircraft to takeoff and land in poor visibility. These systems often have inherent processing delays that can affect a pilot's ability to control an aircraft and a pilot's sense of orientation. The goal of the current study was to determine how much time delay a pilot could tolerate before control was affected, and whether physiological effects would be apparent at the same point. Pilots hovered at a predetermined position in a full flight simulator equipped with a Computer Image Generation (CIG) system and a helmet mounted display. The pilot's visual image was delayed by 67 to 334 milliseconds and varying levels of turbulence were applied to increase the task difficulty. Pilot performance was assessed by collecting objective data on aircraft position error. Handling qualities ratings and reports of physiological symptoms were collected by questionnaire. The results showed that visual time delay increased the variability of position error when as little as 134 ms of delay was encountered. At long delays, sickness symptoms were reported in addition to handing qualities decrements. Turbulence had a minimal effect on performance with long time delays, however it resulted in increased station keeping errors and degraded handling qualities at low delays.

Update rates and fidelity in virtual environments

Interaction is the primary characteristic of a Virtual Environment and update rate is normally taken as an index or measure of the interactivity of the system. The speed of many systems is dictated by the slowest component which is often the Computer Image Generator (CIG). It is common for the workload of the CIG to vary and hence the performance of the system. This paper shows how a variable update rate can produce undesirable results. Two solutions to this problem are presented: service degradation and worst-case. In the case of the CIG, service degradation would require the quality of the image to be reduced such that the time taken never exceeds a given deadline. The worst-case technique works by finding the longest time taken to render any view and then uses that as the deadline for completion. The support of predictive methods is one of several benefits of this approach. An implementation of the worst-case technique is described which takes finer control over the CIG than usual and may be applied to many existing systems with little modification.

Reducing The Latency In Head-Mounted Displays By A Novel Prediction Method Using Grey System Theory

In this paper we propose a novel prediction method for the head motion using Grey System theory, where a 6D tracker is attached to an HMD on a user's head in virtual reality applications. Our prediction method using Grey System Model can greatly reduce the latency by at least one half and reduce image jittering. A system latency below 100 ms or even 50 ins can be achieved, even though without prediction the latency is around 200 ms. Using 6 points in prediction with Grey System Model is currently the best in tracker prediction as we tried from 2 points to 10 points. In order to measure the latency, we also propose a way to measure it in an HMD system precisely and conveniently. During the process, we have implemented four different prototypes respectively on a PC486, a SUN SparcStation10, an SGI IndigoR4000, and a high performance computer image generator. The computation complexity of our prediction method is relatively low and therefore real time requirement is easily met.

Effects of three types of flight simulator visual scene detail on detection of altitude change.

The effects of three types of flight simulator visual scene detail on detection of altitude change were evaluated in three experiments. Across all experiments and with a variety of tasks and display conditions, speed and accuracy of detecting altitude change improved with increases in the density of vertical objects in scenes. Adding detail to individual objects to increase their natural appearance produced no consistent effects on performance. In Experiment 3 complex texture distributed globally on terrain surfaces improved detection of altitude change but did not alleviate the need for high object density. These results indicate that available computer image generator processing capacity would be used more effectively by increasing the density of objects in scenes, rather than by increasing the complexity and detail of individual objects. Complex texture is used more effectively when distributed globally on terrain surfaces, rather than when allocated to individual objects.

Generation and application of high-resolution infrared computer imagery

The generation of high-fidelity simulated infrared imagery requires a unique combination of physical principles and computer image generation technology. At the Georgia Tech Research Institute infrared simulation software has been developed that couples three-dimensional geometric models with geographic databases, infrared radiance prediction models, and computer graphics techniques for image rendering to generate high-resolution synthetic infrared imagery. These features provide a large degree of flexibility to the simulation and allow it to be employed over a wide spectrum of applications. A discussion of a simulation methodology, a review of the GTVISIT scene simulation tool, and a discussion of several applications are given.

RISC design for computer image generation

The paper describes the design and analysis of an instruction set which is applicable to a RISC processor operating as a geometry engine for computer image generation. A prototype LSI implementation of a fixedpoint pipelined RISC is described. It provides a novel technique with DMA channels direct to the register file, a high-performance multiplier and a high-performance divider. The extension of this approach to a VLSI ASIC version using macrocells from a standard cell library is outlined together with a discussion of the testability issues. Results from an LSI implementation and an ASIC simulation are presented.

Computer image generation for flight simulators: the Gabor approach

A formalism for image representation in the combined frequency-position space is presented using the generalized Gabor approach. This approach uses elementary functions to which the human visual system is particularly sensitive and which are efficient for the analysis and synthesis of visual imagery. The formalism is also compatible with the implementation of a variable resolution system wherein image information is nonuniformly distributed across the visual field in accordance with the human visual system's ability to process it. When used with a gaze-slaved visual display system, imagery generated using the techniques described here affords a combination of high resolution and wide field-of-view. This combination is particularly important in high-fidelity, computergenerated, visual environments as required, for instance, in flight simulators.

A three-port/three-access register file for concurrent processing and I/O communication in a RISC-like graphics engine

A novel method that provides extremely fast Input/Output (I/O) data transfer to a RISC-like graphics engine is presented. This method employs a combined two-port / two-access and a three-port / three-access register file used for concurrent processing and I/O data transfer. The three-port cell employed is only 25 % larger than the two-port cell, offering considerable advantages over alternative approaches, such as FIFOs or register banks. This paper discusses some methods that can provide highly fast I/O data transfer in parallel with execution and focuses on the design and implementation of the CMOS-2um register file employed. Index Terms - Computer Architecture, Reduced Instruction Set Computers, VLSI Design, Computer Image Generation, Interprocessor Communication.

Effect of Three-Dimensional Object Type and Density in Simulated Low-Level Flight

Three-dimensional objects placed on simulated terrain surfaces are particularly effective as cues for altitude in simulated low-level flight. To conserve the limited edge processing capacity of computer image generators (CIGs), objects have typically been simple in shape and therefore fairly abstract in appearance. The present investigation sought to determine whether the apparent size of more detailed and familiar appearing objects (e.g., trees and bushes) serves as an additional cue for altitude in simulated low-level flight. Results showed no differences in performance between abstract objects and familiar objects. However, performance did improve with increases in object density, at least for some performance measures. These results suggest that CIG processing capacity may be most effectively utilized by increasing object density rather than individual object detail.

Advanced Simulation

RECOGNISED by authorities throughout the world, flight simulators have in many cases completely replaced the real aircraft for training purposes. Modern techniques are such that a totally realistic and completely risk‐free environment can be created for learning the skills and procedures involved in the operation of expensive aircraft. The leading European manufacturer of these devices and their associated visual systems is Rediffusion Simulation and recent advances of this company have included a distribution computing system and the first flight simualtor instructor facility to combine touch screen technology with microprocessor control. Greatly improved computer image generation combined with a range of visual display systems has ensured extensive applications in both the civil and military fields.

Fast Phong shading

Computer image generation systems often represent curved surfaces as a mesh of planar polygons that are shaded to restore a smooth appearance. Phong shading is a well known algorithm for producing a realistic shading but it has not been used by real-time systems because of the 3 additions, 1 division, and 1 square-root required per pixel for its evaluation. We describe a new formulation for Phong shading that reduces the amount of computation per pixel to only 2 additions for simple Lambertian reflection and 5 additions and 1 memory reference for Phong's complete reflection model. We also show how to extend our method to compute the specular component with the eye at a finite distance from the scene rather than at infinity as is usually assumed. The method can be implemented in hardware for real-time applications or in software to speed image generation for almost any system.

Computer Image Generation of Pre-Columbian Shapes and Patterns

The author describes an artistic project, the objective of which is to recreate Pre-Columbian patterns and objects and give them new meaning using computer image generation techniques. The results are presented in various techniques and formats, including a series of computer-generated photographs, prints and interactive stations.