The generation of real-time 3D graphics scenes normally demands high computational requirements. Several applications can benefit from efficient algorithms for rendering complex virtual environments, such as computer games, terrain visualization, virtual reality and visual simulation. This paper describes an improved view frustum culling method using spatial partitioning based on octrees for 3D real-time rendering. The proposed method is compared against two other approaches. Experiments using four different scenes are conducted to evaluate the performance of each tested method. Results demonstrate that the proposed method presents superior frame rate for all scenes. Parana. His is currently an embedded software developer at Datacom, Curitiba, Brazil. His research interests include computer graphics, computational geometry, computer vision, artificial intelligence.Maikon Cismoski dos Santos received his B.
This paper presents an architecture for 3D virtual environment rendering on the Web applied to computers with limited capacity for graphics processing, which do not have a GPU for rendering the virtual environment. 3D rendering of the scene is done remotely, on demand, and employs 3D warping technique for generating views locally in the client. Thus, the client is able to render new views while waiting for data requested to the server or even when there are problems in the transmission or receiving delay.The proposed architecture allows the virtual environment rendering using residual images, which make the data transmission between client and server more efficient since less information is sent when compared to complete frames rendered from the 3D model. A method for taking advantage of the residual images received by clients is proposed and implemented, improving the quality of the resulting views instead of being discarded. Furthermore, an approach to reducing the quality degradation problem of the views locally rendered in the client is presented.por transmitir menos dados em comparação com os quadros completos renderizados a partir do modelo 3D. Um método para o aproveitamento das imagens de resíduo recebidas pelos clientes foi proposto e implementado, possibilitando que as imagens recebidas pelos clientes sejam utilizadas para o refinamento da qualidade das vistas exibidas, ao invés de serem descartadas. Além disso, uma abordagem para reduzir o problema da degradação da qualidade das vistas renderizadas localmente no clienteé apresentada.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.